Review Article

Management of
Address correspondence to
Dr W. David Freeman, Mayo
Clinic College of Medicine,
4500 San Pablo Rd, Cannaday

Intracranial Pressure 2 East Neurology,

Jacksonville, FL 32224,
[email protected].
W. David Freeman, MD, FSNS, FAAN Relationship Disclosure:
Dr Freeman receives book
royalties from Cambridge
University Press and Oxford
ABSTRACT University Press.
Unlabeled Use of
Purpose of Review: Intracranial pressure (ICP) can be elevated in traumatic brain Products/Investigational
injury, large artery acute ischemic stroke, intracranial hemorrhage, intracranial neo- Use Disclosure:
plasms, and diffuse cerebral disorders such as meningitis, encephalitis, and acute hepatic Dr Freeman reports
no disclosure.
failure. Raised ICP is also known as intracranial hypertension and is defined as a sustained
* 2015, American Academy
ICP of greater than 20 mm Hg. of Neurology.
Recent Findings: ICP must be measured through an invasive brain catheter, typically an
external ventricular catheter that can drain CSF and measure ICP, or through an intra-
parenchymal ICP probe. Proper recognition of the clinical signs of elevated ICP is essential
for timely diagnosis and treatment to prevent cerebral hypoperfusion and possible brain
death. Clinical signs of elevated ICP include headache, papilledema, nausea, and vomit-
ing in the early phases, followed by stupor and coma, pupillary changes, hemiparesis or
quadriparesis, posturing and respiratory abnormalities, and eventually cardiopulmo-
nary arrest.
Summary: Management of elevated ICP is, in part, dependent on the underlying cause.
Medical options for treating elevated ICP include head of bed elevation, IV mannitol,
hypertonic saline, transient hyperventilation, barbiturates, and, if ICP remains refractory,
sedation, endotracheal intubation, mechanical ventilation, and neuromuscular paral-
ysis. Surgical options include CSF drainage if hydrocephalus is present and decom-
pression of a surgical lesion, such as an intracranial hematoma/large infarct or tumor,
if the patient’s condition is deemed salvageable. Future research should continue
investigating medical and surgical options for the treatment of raised ICP, such as
hypothermia, drugs that reduce cerebral edema, and operations aimed at reducing
intracranial mass effect.

Continuum (Minneap Minn) 2015;21(5):1299–1323.

INTRODUCTION MAP is equal to one-third pulse

Intracranial pressure (ICP) is the pres- pressure (PP) plus diastolic blood pres-
sure within the intracranial vault, which sure (DBP): MAP=1/3PP+DBP.
is measured in mm Hg. Normal ICP values PP is equal to systolic blood pressure
are typically 5 mm Hg to 15 mm Hg. In (SBP) minus DBP: PP=SBPYDBP.
normal states, ICP typically mirrors jug- Assuming a normal blood pressure
ular venous pressure, which is about of 120/80 mm Hg, MAP calculated from
5 mm Hg to 8 mm Hg. Pathologic states the equations above will be approxi- Supplemental digital content:
Videos accompanying this ar-
(Table 3-1) cause intracranial hyperten- mately 93 mm Hg. Assuming ICP is nor- ticle are cited in the text as
mal (eg, 5 mm Hg), CPP (93 minus 5) Supplemental Digital Content.
sion, conventionally defined as sustained Videos may be accessed by
elevations of ICP greater than 20 mm Hg. would be approximately 88 mm Hg. In clicking on links provided in the
pathologic states, as ICP increases, CPP HTML, PDF, and app versions of
Intracranial hypertension can reduce this article; the URLs are pro-
cerebral perfusion pressure (CPP). CPP approaches zero, which can cause brain vided in the print version. Video
ischemia and neuronal death if not legends begin on page 1320.
is equal to mean arterial pressure (MAP)
minus ICP: CPP=MAPYICP. rapidly corrected. The protective reflex

Continuum (Minneap Minn) 2015;21(5):1299–1323 www.ContinuumJournal.com 1299

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

 Intracranial Pressure

TABLE 3-1 Etiologies and Pathophysiology of Elevated Intracranial Pressure

Basic Categories of Intracranial

Pressure Elevation Examples of Intracranial Pressure Pathophysiology
Traumatic brain injury Diffuse axonal injury, diffuse or focal cerebral edema,
intracranial hemorrhage, or contusions with mass effect
Large vessel ischemic stroke (eg, internal Large-volume cytotoxic cerebral edema with mass effect causing
carotid artery or middle cerebral secondary cerebral herniation and vascular occlusion of
artery occlusion) nearby arteries
Intracranial hemorrhage Mass effect within closed intracranial vault (eg, subdural
hematoma, intracerebral hematoma, intraventricular
hemorrhage, subarachnoid hemorrhage with or
without hydrocephalus)
Hydrocephalus Noncommunicating: also referred to as obstructive (mass effect or
lesions that obstruct the major CSF ventricular pathways or drainage)
Communicating: blockage of arachnoid granulations or microscopic
CSF drainage from cells (eg, bacterial meningitis, subarachnoid
hemorrhage) or high protein (eg, Guillain-Barré syndrome)
Diffuse cerebral edema Acetaminophen overdose causing acute hepatic failure
with hyperammonemia, acute-on-chronic hepatic failure,
encephalitis/cerebritis or meningitis, from vasogenic or other
forms of cerebral edema
Jugular venous obstruction or elevated Jugular venous thrombosis (central line related or hypercoagulable
right-sided cardiac venous pressure state), superior vena cava syndrome, severe congestive heart
failure, or chronic obstructive pulmonary disease with clinically
significant papilledema
Brain neoplasms Mass effect, vasogenic edema
Idiopathic edema Pseudotumor cerebri, pathophysiology remains speculative, main risk
factors include obesity and female sex
CSF = cerebrospinal fluid.

KEY POINT response to elevated ICP causes changes cranial content must remain constant
h Normal intracranial in blood pressure and heart rate as orig- and, therefore, any change in one of the
pressure is typically less inally described by Harvey Cushing.1 intracranial components must be com-
than 15 to 20 mm Hg. Many patients with raised ICP have ele- pensated by a reciprocal change in the
vated blood pressure as a result of this volume of another component. Math-
compensatory reflex to try to maintain ematically, the intracranial pressure-
adequate CPP. volume relationship can be represented
as follows: VT=Vb+Vcsf+Vvasc.
HISTORY OF THE MONRO-KELLIE In this equation, VT refers to total in-
DOCTRINE tracranial volume (1700 mL fixed in adult-
The intracranial pressure-volume relation- hood), Vb refers to brain parenchymal
ship was initially described by Alexander volume, which is about 1400 mL, Vcsf
Monro secundus2Y5 and expanded on refers to CSF intracranial volume (approx-
by George Kellie in what is known today imately 30 mL in ventricles), and Vvasc
as the Monro-Kellie doctrine. This doc- refers to circulatory volume (arteries and
trine states that the volume of the intra- veins), which is approximately 150 mL.6

1300 www.ContinuumJournal.com October 2015

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

 KEY POINT
In pathologic states, a new intracra- causing brain damage or ischemia and h Cerebral perfusion
nial volume (Vx) is added to the intracra- infarction (Figure 3-1). pressure is compromised
nial compartment: VT = Vb + Vcsf + There are three fossa or recesses in- by intracranial pressure
Vvasc+ Vx (Table 3-1). To accommo- tracranially: (1) the posterior fossa is elevation, as expected by
date this new intracranial volume, one separated by the tentorium cerebelli, the following equation:
of the other components must be dis- which is a thick dural membrane that cerebral perfusion
placed accordingly to maintain constant contains most of the brainstem that exits pressure is equal to mean
ICP. Most commonly, the first intracra- the foramen magnum; (2) the anterior arterial pressure minus
nial component to be displaced is CSF fossa is where the cribriform plate and intracranial pressure.
(Vcsf), which egresses from the ven- frontal lobes reside; and (3) the mid-
tricles or cerebral convexities through dle fossa holds the temporal lobes and
the arachnoid granulations into the is medially adjacent to the border of the
dural venous sinuses. However, this rep- tentorium cerebelli. When an intracra-
resents a temporary compensation. Fur- nial mass effect occurs inside of these
ther additions of intracranial volume fossa (eg, temporal lobe hematoma
may exhaust intracranial reserve mech- with mass effect), it can encroach and
anisms, leading to a rise in ICP as well extend into the tentorium cerebelli
as potential compression of other intra- causing displacement of adjacent intra-
cranial structures, such as brain paren- cranial structures, such as the third cra-
chyma (Vb) or vascular structures (Vvasc), nial nerve, and subsequent ipsilateral

FIGURE 3-1 Herniation. The figure demonstrates a subdural hematoma

causing increased intracranial pressure and mass effect
and compression on the third cranial nerve as well as
microvascular compression on perforating deep brain parenchymal arteries
and arterioles, which can lead to vascular brain and brainstem injury.
ACA =anterior cerebral artery; PCA = posterior cerebral artery.
Used with permission of Mayo Foundation for Medical Education and Research. All rights reserved.

Continuum (Minneap Minn) 2015;21(5):1299–1323 www.ContinuumJournal.com 1301

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

 Intracranial Pressure

KEY POINT
h Clinical signs of pupil dilatation. If mass effect extends exhausted intracranial compliance.
herniation include diagonally across the brainstem to cause These sustained elevations, known as
deteriorating mental it to abut the contralateral tentorium plateau waves (or A waves), were first
status, dilatation of one cerebelli (dural edge), this can cause described by Lundberg and can last
or both pupils, posturing contralateral third nerve injury and pu- for minutes to an hour or longer un-
of one or later both sides, pil dilatation (ie, Kernohan notch phe- less some intervention is instituted to
and eventual cessation nomenon). The initial mechanisms of alleviate them.3,7,8 Shorter, self-limited
of respiration. reserve include CSF egress into the B waves can also occur, which may her-
ventricular system and compression of ald the appearance of A waves. Thus,
space between cortical sulci or CSF prompt intervention is necessary to pre-
spaces. After these mechanisms are vent brain damage when ICP begins to
exhausted, brain tissue itself becomes rise. During these plateau waves, CPP
compressed and shifts in paths of least can become progressively more com-
intracranial resistance, causing me- promised, and patients may develop
chanical movement of brain tissue, as signs of herniation. Timely treatment of
described above, called cerebral her- elevated ICP is needed to prevent a low
niation, or brain tissue moving across CPP crisis and to halt the progressive
one side of the dural fossa into another downward spiral in CPP toward zero.
anatomic space (eg, uncal herniation is
the temporal lobe uncus shifting from MEASUREMENT OF INTRACRANIAL
middle fossa into tentorium and at- PRESSURE FROM INTRACRANIAL
tempting to move medially downward AND LUMBAR LOCATIONS
toward the lower posterior fossa) (shown The ICP zero point is defined as the
in Figure 3-1 by the lower black arrow center of the head or at the level of
near the compressed posterior cere- the foramen of Monro,9Y11 which is
bral artery). Subfalcine herniation occurs anatomically close to the tragus of the
when part of the frontal lobe moves outer ear (Figure 3-2 and Figure 3-3).
under the falx cerebri to the contra- When a patient is lying on his or her
lateral side (shown by the upper black side and the center of the head is in
arrow in Figure 3-1), which can cause line with the center of the spinal needle
compression of the anterior cerebral or catheter, intrathecal pressure is equiv-
artery blood supply and potential cere- alent to ICP, assuming there is no
bral infarction from mechanical occlu- blockage of CSF from cranial to caudal
sion of these arteries (ipsilaterally and structures. This is why a lumbar punc-
contralaterally if severe). Tonsillar her- ture should be performed with the pa-
niation occurs when the cerebellar ton- tient lying on his or her side (lateral
sils herniate downward through the decubitus position) so that the head
foramen magnum. Diencephalic her- is in line with the spine. Performing a
niation occurs when the diencephalon lumbar puncture in the sitting position
herniates downward through the ten- makes measuring ICP problematic and
torium into the posterior fossa, typi- may be inaccurate unless the tubing
cally from bilateral cerebral edema or system is long enough to extend above
similar bilateral mass effect. External the tragus. Most lumbar puncture kits do
herniation occurs when part of the not have sufficient manometric tubing
calvarium is removed (surgically or by to accomplish this. Another important
trauma) and the brain herniates out- consideration in measuring CSF open-
side the intracranial vault. ing pressure (as an estimate of ICP) in
ICP monitoring demonstrates sus- the lateral decubitus position is that the
tained elevations of ICP in patients with tubing manometer is usually measured
1302 www.ContinuumJournal.com October 2015

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

FIGURE 3-2 Intracranial pressure (ICP) monitoring devices. Different types of ICP monitoring devices, including epidural and
subarachnoid (hardly used) devices and intraparenchymal and intraventricular methods. An external ventricular
drain is considered the gold standard for ICP measurement.
Used with permission of Mayo Foundation for Medical Education and Research. All rights reserved.

in mm and cm of H2O. Normal ICP is typi- the cranium, the shape of the ICP wave- KEY POINT
cally less than 15 mm Hg to 20 mm Hg. form has three distinct waves, some- h While elevated
Conversion from cm H2O to mm Hg re- times called P1, P2, and P3. Case 3-1 intracranial pressure
may be suspected on
quires division by a factor of 1.35. For illustrates these ICP distinct waves. As
clinical grounds,
example, 27 cm H2O is equal to 20 mm Hg. additional intracranial volume is added,
intracranial pressure
Similar to ICP measured by lumbar punc- ICP increases precipitously (Figure 3-7). requires measurement
ture, lumbar drain catheters can mea- Case 3-2 illustrates a case of a grossly with an invasive probe
sure ICP as long as the zero point of the abnormal ICP waveform from this rela- or external ventricular
system is set at the tragus of the ear, tionship. This mathematical and graphical drain to confirm.
similar to an external ventricular drain relationship of ICP change from intra-
(EVD) (Figure 3-4). cranial volume changes has been termed
compliance (C) throughout the litera-
INTRACRANIAL PRESSURE ture,13 but, in fact, elastance (E) is the
WAVEFORM INTERPRETATION correct term for this mathematical re-
AND CORRELATIONS TO THE lationship.14 However, since compliance
PRESSURE-VOLUME RELATIONSHIP and elastance are mathematically re-
The ICP waveform is important because ciprocal (ie, E = 1/C), both terms are
its morphology provides clues to the often used (Figure 3-7). Compliance can
pressure-volume relationship within be conceptualized as how stretchy an
the intracranial vault (Figure 3-512). object is (ie, volume change) due to
When ICP is normal and a compensated changes in pressure (ie, C = dV/dP,
pressure-volume relationship exists within where dV equals change in volume and

Continuum (Minneap Minn) 2015;21(5):1299–1323 www.ContinuumJournal.com 1303

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

 Intracranial Pressure

FIGURE 3-3 External ventricular drain drainage and setup. The figure shows the intraventricular catheter (ie, external ventricular
drain [EVD]), which is hooked up to a drainage (closed-contained, sterile) system.
Used with permission of Mayo Foundation for Medical Education and Research. All rights reserved.

KEY POINT dP equals change in pressure), whereas shown in videos and discussed in the
h The pressure-volume elastance represents a pressure change cases of this article.
relationship is known as
(ie, ICP change) in relation to a given
compliance or elastance, BEDSIDE MANEUVERS FOR
depending on whether
volume change (E = dP/dV). Elastance
can also be thought of as recoil resis- ASSESSMENT OF INTRACRANIAL
pressure or volume is on
tance. None theless, in states of low COMPLIANCE AND ELASTANCE
the X or Y axis since they
are inversely related. intracranial compliance or high elas- Bedside maneuvers can help make in-
tance, the ICP waveform changes from ferences about the intracranial com-
a distinct shape with three waves to a pliance and elastance relationship. A
more triangular, spikelike appearance simple bedside maneuver in comatose
(Figure 3-5).12,13,15 Therefore, the ICP patients is jugular vein compression
waveform provides insight into intra- (either unilateral or bilateral) while
cranial elastance/compliance and is im- monitoring ICP. This is performed by
portant to recognize. Two examples of simply applying gentle manual pressure
abnormal ICP waveform morphology are over each jugular vein of the neck

1304 www.ContinuumJournal.com October 2015

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

FIGURE 3-4 Lumbar drain setup and zero for intracranial pressure. Patient is shown in the left lateral decubitus position, with
the zero of the drainage system in line with the head (zero), indicated by the red dashed line. When a patient is
sitting with head of bed at 30 degrees, the zero of the drainage system should be kept in line with the tragus,
which is approximately at the level of the foramen of Monro or the intracranial zero point.
Used with permission of Mayo Foundation for Medical Education and Research. All rights reserved.

while observing ICP under continuous CONT/A152). The degree of ICP eleva-
monitoring. This is only necessary for tion in patients with high elastance/
comatose patients who cannot report low compliance intracranial states can
symptoms of high ICP, such as head- be striking and provide useful infor-
ache, or perform a voluntary Valsalva mation to guide treatment.
maneuver. With jugular vein compres- A second maneuver is placing the
sion, ICP is expected to increase simi- patient’s head of bed at 0 degrees
larly to the rise seen with Valsalva ( head flat test) from the common 30
maneuver in awake patients. After re- to 40 degree elevation used to reduce
moval of jugular venous compression, ventilator-associated pneumonia. The
ICP should decrease (Supplemental ICP is first assessed with the head of bed
Digital Content 3-1, links.lww.com/ at 30 to 40 degrees and then reassessed

Continuum (Minneap Minn) 2015;21(5):1299–1323 www.ContinuumJournal.com 1305

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

 Intracranial Pressure

KEY POINT
h The intracranial pressure
waveform appearance
provides clues to the
state of normal
or abnormal
intracranial compliance.

FIGURE 3-5 Intracranial pressure waveform in normal (upper line) and abnormal (lower line)
intracranial pressure-volume states.
Reprinted with permission from Chesnut RM, Marshall LF. Neurosurg Clin N Am.12 B 1991 Elsevier Inc.

when the head of bed is at 0 degrees. ondary brain injury.16,17 Another test
In patients with an EVD, this will re- has been described involving the injec-
quire clamping of the EVD and rezeroing tion of a small volume of sterile saline
after head position change. Intraparen- into either a lumbar or ventricular cath-
chymal continuous ICP monitors do not eter followed by mathematical analy-
require rezeroing with changes in head sis of the patient’s elastance curve.18
position. Increased ICP with the head However, this remains a research tool
of bed at 0 degrees indicates poor in- at present.
tracranial compliance and demands
CEREBRAL PERFUSION
caution, particularly when planning trans- PRESSURE AND CEREBRAL
portation outside of the intensive care BLOOD FLOW: PRESSURE
unit or testing with the head in a flat AUTOREGULATION
position (eg, for CT scanning).16 For pa- CPP values are mathematically depen-
tients with refractory intracranial hy- dent on ICP via the equation, CPP is
pertension, being laid flat for the 20 to equal to MAP minus ICP: CPP=MAPYICP.
30 minutes necessary to obtain a CT or Additionally, cerebral blood flow
MRI without continuous ICP monitor- (CBF) values are dependent on CPP.
ing can potentially lead to CPP crisis Therefore, the equation for CBF is impor-
states (less than 60 mm Hg) and sec- tant to know in the clinical management

1306 www.ContinuumJournal.com October 2015

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

 Case 3-1
A 45-year-old man presented with an aneurysmal subarachnoid hemorrhage with hydrocephalus
and had an external ventricular drain (EVD) placed for treatment of communicating hydrocephalus
(Figure 3-6). The patient was awake and interactive while undergoing a clamping trial of his EVD system
(Supplemental Digital Content 3-1, links.lww.com/CONT/A152), which demonstrated an intracranial
pressure (ICP) waveform with clustering of P1 to P3 waves, suggesting a mildly abnormal intracranial
compliance/elastance state. The effects of respiration were seen as undulations of the ICP
(purple tracing in the video) during inspiration, which decreased ICP by pulling jugular venous blood
from the head into the right atrium owing to the negative intrathoracic pressure. Also, the ICP pressure
waveform showed increases in ICP while the patient’s deltoid strength was tested because testing
caused the patient to perform a Valsalva maneuver. The Valsalva maneuver causes the glottis to
close, and the increased intrathoracic pressure increases jugular venous pressure and causes a
transient increase in ICP.

FIGURE 3-6 Axial noncontrast head CT of the patient in Case 3-1 showing subarachnoid
hemorrhage blood in the basal cisterns and fourth ventricle (A), as well as mild
hydrocephalus and external ventricular drain placement and blood layering out in
the posterior horns of the lateral ventricles (B).

Comment. This case illustrates the importance of visual inspection of the ICP waveform and
helps provide information about the intracranial pressure-volume relationship and communicating
hydrocephalus, as this patient experienced after subarachnoid hemorrhage.

of patients with ICP elevation. CBF is range of CPP values by the mechanism
equal to CPP divided by cerebrovascular of cerebrovascular autoregulation. In
resistance (CVR): CBF=CPP/CVR. normal states, cerebrovascular autoreg-
The above equation is also similar ulation varies commensurately with CPP
to Ohm law: voltage (V) is equal to cur- values (Figure 3-9).9 In other words,
rent (I) times resistance (R) (V = IR), or as CPP increases, CVR increases, and
solving for I, where I equals V divided vice versa. Contrastingly, autoregulation
by R (I = V/R).19 In normal circum- is an energy-dependent process that
stances, CBF is held constant across a occurs at the arteriolar level requiring

Continuum (Minneap Minn) 2015;21(5):1299–1323 www.ContinuumJournal.com 1307

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

 Intracranial Pressure

KEY POINTS
h Cerebrovascular
autoregulation is often
disturbed after brain
injury, causing cerebral
blood flow to have a
linear dependence on
cerebral perfusion
pressure, which creates
risk for brain tissue
oligemia and ischemia at
lower cerebral perfusion
pressure thresholds
(cerebral perfusion
pressure of less than
65 mm Hg).
h Chronically hypertensive
patients live at a higher
zone of cerebral
perfusion pressure
values than patients FIGURE 3-7 Change in pressure divided by change in volume (dP/dV) elastance curve. As
with normal increasing intracranial volume is added, intracranial pressure begins to rise
precipitously, and this, in turn, is reflected on the shape of the intracranial
blood pressure. pressure waveform. Elastance (E) = dP/dV, whereas compliance is the opposite (dV/dP), and
the terms are often used interchangeably. Therefore, both terms are used appropriately in this
figure depending on the part of the curve described.
ICP = intracranial pressure.

adenosine triphosphate (ATP) within an ICP monitor is critical for ICP mea-
the vessel wall to either constrict or surement and CPP calculation to pre-
dilate. However, in acute brain injury vent secondary ischemia to the brain
such as stroke, traumatic brain injury from low CPP and CBF states. Methods
(TBI), or hemorrhage, the ability to for assessing the presence or absence
autoregulate can be lost, and CBF be- of autoregulation exist but typically re-
comes linearly passive to CPP values quire the use of multimodality moni-
(Figure 3-9). This linear relationship toring,20Y23 which may not be practical
between CPP and CBF values presents for most centers.
a challenge to patients with brain injury
since low CPP values may drop below THE MANAGEMENT OF THE
the critical CBF threshold of 20 mL/ PATIENT WITH ELEVATED
100 g/min, which is the ischemic thresh- INTRACRANIAL PRESSURE
old (Figure 3-9). Therefore, patients with The clinical signs and symptoms of ele-
brain injury may develop two different vated ICP include headache, nausea,
pathophysiologic problems: (1) com- vomiting, and papilledema in awake
promised CPP from ICP elevation with patients, which may progress to stu-
or without preserved autoregulation por and coma over time. Patients who
and (2) loss of autoreguation, which present with acute stupor or coma have
makes CBF passively dependent on CPP a rapid spike in ICP leading to pupillary
values. Either process can lead to sec- dilatation, flexor or extensor postur-
ondary forms of brain injury and ische- ing, and abnormal respiratory function
mia. In these situations, placement of progressing to cardiopulmonary arrest

1308 www.ContinuumJournal.com October 2015

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

 Case 3-2
A 47-year-old woman experienced severe subarachnoid and bifrontal intraparenchymal hemorrhages
from a ruptured anterior communicating artery aneurysm (Figure 3-8). The patient was intubated,
sedated, under neuromuscular blockade, and mechanically ventilated because of refractory intracranial
hypertension. The patient demonstrated a noncompliant (ie, low compliance, high elastance)
intracranial pressure (ICP) waveform (Supplementary Digital Content 3-2, links.lww.com/CONT/A153)
that had a frank triangular appearance and no distinct P waves. The ICP monitor demonstrated
periods where the external ventricular drain (EVD) was opened to drain a few milliliters of CSF, which made
the digital ICP waveform go flat transiently while CSF emptied into the CSF collection chamber
(not shown) and was then closed again from CSF drainage to transduce ICP and show the waveform.
This was essentially the opposite of the CSF infusion test described in the text because it allowed CSF to
drain out. In other words, the patient’s elastance curve moved from the right side of the graph to the
lower left (Figure 3-7), which indicated a reduction in elastance (increase in compliance) from CSF
removal. However, despite CSF drainage, the ICP waveform still demonstrated an abnormal ICP
morphology, indicating that the intracranial pressure-volume state remained abnormal. This
underscored the importance of the ICP waveform and the insight it provided into the intracranial
pressure-volume state independent of the numeric value of ICP measurement.

FIGURE 3-8 Imaging of the patient in Case 3-2. Axial noncontrast head CT (A, B) demonstrating bifrontal parenchymal
intraventricular hemorrhage from an anterior communicating artery aneurysm with communicating
hydrocephalus and global cerebral edema. The arrow in panel B demonstrates the mass effect of the left-sided
hemorrhage with left to right midline shift of the falx cerebri. The subsequent scan (C) shows left frontal craniectomy with
bone flap replacement, evacuation of the hematomas, and external ventricular drain placement.

Comment. This case illustrates how bedside ICP waveform analysis and CSF drainage can help
provide insight into the intracranial pressure-volume relationship and physiology. In this patient’s case,
the ICP values and waveform analysis helped make inferences about the patient’s intracranial
compliance/elastance state before de-escalating treatments such as neuromuscular blockade. This patient
eventually had improved intracranial compliance/elastance due to bifrontal edema resolution and had
de-escalation of therapies for refractory ICP (ie, neuromuscular blockade and eventually sedation) while
maintaining an as-needed use of hypertonic saline for intermittent spikes in ICP above 20 mm Hg.

and death unless there is an interven- with a drop in CPP and lose conscious-
tion to reduce ICP. Some patients, such ness, followed by a drop in ICP after
as those with a ruptured intracranial the aneurysm stops bleeding and CPP
aneurysm with subarachnoid hemor- returns to more normal levels. This is
rhage, have an immediate spike in ICP why some patients lose consciousness

Continuum (Minneap Minn) 2015;21(5):1299–1323 www.ContinuumJournal.com 1309

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

 Intracranial Pressure

KEY POINT
h Management of
intracranial pressure is
typically performed in a
multifaceted approach
including the ABCs
(airway, breathing, and
circulation) of
cardiopulmonary support,
head and neck posture
optimization, head of
bed elevation, intracranial
pressure monitoring
(with cerebral perfusion
pressure optimization),
and use of medical
therapies to reduce
intracranial pressure
including mannitol or
hypertonic saline.

FIGURE 3-9 Cerebral blood flow and cerebral perfusion pressure relationship of
autoregulation. Cerebral perfusion pressure is shown on the X axis, while cerebral
blood flow is shown on the on Y axis. In brain injury, autoregulation is
disturbed and cerebral blood flow becomes linearly dependent on cerebral perfusion pressure.
CPP = cerebral perfusion pressure; MAP = mean arterial pressure;
ICP = intracranial pressure.
9
Modified with permission from Rose JC, Mayer SA, Neurocrit Care. link.springer.com/article/10.1385%2FNCC%3A1%
3A3%3A287 B 2004, Humana Press Inc.

at the time of intracranial aneurysm Patients who present with acute

rupture. These patients may later suc- stupor or coma should be emergently
cumb to delayed increased ICP from managed with cardiopulmonary sup-
communicating or obstructive hydro- port, the ABCs (airway, breathing, and
cephalus. While elevated ICP may be circulation), to maintain oxygenation
suspected on clinical grounds alone or (oxygen peripheral saturation level of
in conjunction with acute imaging find- greater than 92% to 95% in most
ings, an intracranial device to measure cases), sufficient blood pressure
ICP can confirm elevated ICP. Papille- (MAP of greater than 70 mm Hg), and
dema, when present on examination, airway protection. Patients with a Glas-
may indicate that ICP has been greater gow Coma Scale (GCS) score of 8 or
than 20 mm Hg. However, this physical lower should typically receive rapid se-
examination finding may not be pres- quence intubation for airway protection.24
ent within the first few hours of acute These cardiopulmonary support ABCs
ICP elevations and may only become should be performed in parallel with
evident later. Papilledema may also re- the initial neurologic examination to
present a chronic finding in some pa- ensure the patient is stable enough to
tients, such as those with idiopathic undergo additional testing, such as
intracranial hypertension (pseudotumor neuroimaging. A general approach
cerebri). Also, some patients with in- for managing ICP is provided in the
creased ICP may never develop papil- ICP algorithm in Figure 3-10, but in
ledema because of anatomic reasons. each case individualization is advised

1310 www.ContinuumJournal.com October 2015

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

FIGURE 3-10 Algorithm for elevated intracranial pressure. A multifaceted approach should be used when managing a patient
with increased intracranial pressure.
CPP = cerebral perfusion pressure; CT = computed tomography; EVD = external ventricular drain;
GCS = Glasgow Coma Scale; ICP = intracranial pressure; ICU = intensive care unit.
a
Neurosurgical operable lesions include symptomatic hydrocephalus, posterior fossa hematoma 3 cm in size or greater, and
symptomatic subdural hematoma in patients with a reasonable chance for survival and relatively good neurologic outcome.

based on various factors. Additionally, (Table 3-1). If intracranial hemorrhage,

an outline of methods to reduce ICP hydrocephalus, or intracranial neoplasm
is provided in Table 3-2.25Y28 Severe is recognized on brain imaging, emer-
and refractory ICP elevations typically gent neurosurgical consultation is
indicate the severity of the primary ill- advised for possible surgical decom-
ness and may portend worse outcomes, pression, ICP monitor placement, or
especially in TBI.29 In this sense, refrac- ventricular catheter placement for hy-
tory ICP is a ‘‘biomarker’’ for severity of drocephalus as recommended by the
illness beyond the GCS designation.29Y32 American Heart Association/American
In patients with acute alterations of Stroke Association guidelines for pa-
consciousness or localizing neurologic tients with subarachnoid hemorrhage
signs such as acute anisocoria (greater and intracranial hemorrhage.33Y35
than 1 mm), brain imaging is warranted Patients with bilateral fixed and di-
to evaluate the causes of elevated ICP. lated pupils or unresponsive corneal
Typically a noncontrast CT is used in reflexes on examination typically have
emergency situations and ordered im- a poor outcome after intracranial hem-
mediately to help determine the under- orrhage, even with emergent neuro-
lying cause of the acute ICP elevation surgical decompression. 36 Recent

Continuum (Minneap Minn) 2015;21(5):1299–1323 www.ContinuumJournal.com 1311

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

 Intracranial Pressure

TABLE 3-2 Overview of Methods to Reduce Intracranial Pressure

Treatment Pros Cons and Monitoring

Head and neck position Easy bedside intervention with modest This is an often overlooked area in
and head of bed elevation to significant effects depending on the intracranial pressure (ICP), since
head and neck position. Head of bed overextension or extreme neck flexion
should be at 30Y45 degrees. Higher and neck rotation can occlude jugular
head of bed elevations (945 degrees) veins and increase ICP. Also, in intubated
reduce cerebral perfusion pressure. patients, endotracheal tube holder ties
can be too tight, causing jugular vein
compression. This treatment requires
nurse and respiratory therapist education.
Mannitol 20%Y25% Safe administration through a Osmotic diuretic; patients should
(0.25Y1.5 g/kg IV piggyback peripheral IV route makes mannitol be monitored for volume loss from
or bolus) a good option in the emergency diuresis (large urine output 91 L) and
department and makes it the common subsequent hypotension as well as
front-line agent for ICP emergencies. electrolyte replacement.
The dose chosen depends on the Serum sodium and serum osmolarity
severity of ICP elevation. For example, monitoring (keep sodium G155Y160
in patients with temporal lobe mmol/L and osmolarity G320 mOsm/kg).
hematoma with acute uncal herniation Avoid in patients with advanced renal
and pupillary dilatation, 1.5 g/kg IV disease because it may not be excreted.
bolus can be used to reverse Electrolyte monitoring and replacement
herniation until surgical evacuation (eg, hypokalemia, hypophosphatemia).
is performed. For patients with milder
ICP elevations, smaller doses such as If mannitol gets too cold it can
0.25 g/kg IV piggyback can be used precipitate like snowflakes; until
as needed. dissolved or warmer, it can be infused
with an IV filter along the IV tubing.
In emergency settings, absolute
doses of 25 g, 50 g, 75 g, or 100 g IV
bolus can be given immediately from
bags that contain 100 g total, which
prevents delays in calculating the
weight-based dose (eg, for patients
acutely herniating, give the
100 g bag immediately).
Hypertonic saline Equimolar doses to mannitol are Most centers require a central line
3% infusion similarly effective on ICP response. for 3% hypertonic saline or higher to
Hypertonic saline provides prevent thrombophlebitis via peripheral
14.6% bolus hyperosmolar state and increased IV for use. Risk includes overshoot
(24 mL or 48 mL)
circulating blood volume compared hypernatremia and hyperchloremic
23.4% bolus to mannitol, which causes volume acidosis. Caution is advised in those with
(15 mL or 30 mL) depletion. Hypertonic saline may subacute or chronic hyponatremia to
be preferable in certain patient avoid central pontine myelinolysis.
populations, such as subarachnoid Monitoring is similar to mannitol.
hemorrhage with vasospasm, in
which volume depletion may
compromise cerebral perfusion.

Continued on page 1313

1312 www.ContinuumJournal.com October 2015

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

TABLE 3-2 Overview of Methods to Reduce Intracranial Pressure Continued from page 1312

Treatment Pros Cons and Monitoring

Sedation and analgesia Pain control and sedation may reduce Oversedation may depress respiratory
(eg, propofol or midazolam, ICP by a modest degree by reducing drive and require intubation and
and fentanyl infusions) Valsalva maneuver and jugular venous mechanical ventilation and also
pressure elevation. obscures the neurologic examination.
Propofol and benzodiazepines also
have antiseizure properties.
Neuromuscular paralysis Neuromuscular paralysis prevents Requires intubation, mechanical ventilation;
(curare-derived drugs [eg, coughing, which may cause ICP spikes loss of neurologic examination aside
cisatracurium, vecuronium]) from Valsalva maneuver. from pupillary function.
Barbiturates (eg, Reduce brain metabolism Requires intubation and mechanical
pentobarbital or thiopental) (put brain to ‘‘sleep’’) to lower ICP. ventilation, and most perform
EEG monitoring, titrating to
Antiseizure side benefit.
suppression-burst pattern.
Reduces cardiac output via negative
inotropy, which may require
vasopressor support.
Longer use leads to accumulation in
adipose tissue and may impede a reliable
neurologic examination for prognostication
or brain death declaration.
Hypothermia (body Reduces brain metabolism similar Effective for ICP control25 but does not
temperature 32-CY34-C to barbiturates. change traumatic brain injury outcome
[89.6-FY93.2-F]) per National Acute Brain Injury Study
trials.26 Longer use (48 to 72 hours) of
hypothermia in patients with traumatic
brain injury was described in one study27
with favorable outcomes but remains
controversial.26Y28
Increased risk of infection. Changes in
drug metabolism prolong the effects
of sedatives.
Neurosurgical interventions Depends on the lesion, but includes Surgical risks of procedure and device
external ventricular drain placement, implantation (ventriculostomy infection
which can be diagnostic and therapeutic or meningitis).
(measure ICP and drain CSF volume to Only applicable to certain pathologies,
reduce ICP), craniotomy and mass lesion such as hydrocephalus, subdural
resection (eg, intracranial hemorrhage hematoma, posterior fossa intracranial
or brain tumor). hemorrhage of 93 cm and decompressive
hemicraniectomy in some patients with
large middle cerebral artery infarction.
Typically lifesaving procedure but may
not change or improve functional
outcome of survivors.

CSF = cerebrospinal fluid; EEG = electroencephalogram; IV = intravenous.

Continuum (Minneap Minn) 2015;21(5):1299–1323 www.ContinuumJournal.com 1313

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

 Intracranial Pressure

guidelines for palliative care of such ment.’’ Prolonged or empiric aggressive

patients have been proposed by the hyperventilation (eg, PaCO2 of approxi-
American Heart Association/American mately 25 mm Hg range) resulted in
Stroke Association.37 In patients with worse outcomes of patients with TBI in
TBI, the Brain Trauma Foundation in one clinical trial.39 Another recent inter-
collaboration with the American As- national consensus guideline suggests
sociation of Neurological Surgeons and using ICP monitoring in patients with a
the Congress of Neurological Surgeons GCS score of 8 or lower only when CT
suggest, along with more recent con- imaging abnormalities are present.40
sensus guidelines,29 that patients with
TBI who have a GCS score of 8 or less TYPES OF INTRACRANIAL
and CT imaging abnormalities should PRESSURE MONITORING
be fitted with an ICP monitor.38 A re- The type of intracranial device used to
cent randomized study has, however, monitor ICP (Figure 3-2) is typically
created some controversy around ICP decided by the neurosurgeon based
monitoring in patients with severe TBI on relative risks and whether CSF
(ie, a GCS score of 8 or less). This study drainage is needed. The types of ICP
found no difference in the outcome monitors are basically divided between
of patients with severe TBI when the those that allow ventricular CSF drain-
management was guided by ICP mon- age and those that do not ( parenchy-
itoring or simply by findings on physi- mal probes). Table 3-3 10,12,33,35,40Y44
cal examination and brain imaging.11 provides a high-level summary of
However, many questions were raised the benefits, risks, and evidence or
in the non-ICP monitored arm of the guidelines for different ICP monitor-
study, such as the use of hyperventila- ing devices, with the main benefits or
tion as part of the ‘‘empiric ICP treat- risks highlighted.

a
TABLE 3-3 Benefits, Risks, and Evidence for Use of Intracranial Pressure Monitors

Method Benefits Risks Evidence or Guidelineb

Ventricular intracranial Gold-standard intracranial Invasive, external ventricular Level II: Traumatic brain
pressure monitor pressure, cerebral drainYtract bleeding injury,10 Glasgow
perfusion pressure, in (2%Y10%),12,41 intermittent Coma Scale score of less than 8,
vivo calibration, global intracranial pressure, abnormal CT10)
ventriculitis/meningitis Level I: Symptomatic
(5%Y20%)41Y43 hydrocephalus (American Heart
Association/American Stroke
Association subarachnoid
hemorrhage and intracerebral
hemorrhage guidelines33,35,44)
Parenchymal intracranial Continuous intracranial Invasive, infection, bleeding, Similar to ventricular, but
pressure monitor pressure/cerebral drift after 7Y10 days, ventricular ranked first (Brain
perfusion pressure,12,40 no CSF drainage Trauma Foundation/American
global and regional Association of Neurological
intracranial pressure Surgeons/Congress of
Neurological Surgeons10)
a
Italicized text signifies major complications or issues among the benefits and risks of intracranial pressure monitors.
b
Levels of evidence proposed by American Heart Association guidelines.

1314 www.ContinuumJournal.com October 2015

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

 KEY POINTS
MEDICAL THERAPY FOR (Na + K)) + (BUN/2.8) + (glucose/18). h For patients with
INTRACRANIAL HYPERTENSION An osmolal gap greater than 15 is ab- refractory intracranial
Over the past century, many agents normal and suggests a higher than pressure, typical
that reduce ICP have been studied, normal amount of substances in the treatments include
including glycerol, furosemide, 20% blood such as mannitol, ethanol, ethyl- sedation, neuromuscular
mannitol, hypertonic saline (3% and ene glycol, or methylene alcohol. paralysis, mechancial
higher concentrations), barbiturates, Mannitol also has a reflection coef- ventilation, barbiturates,
and indomethacin.44 However, for the or hypothermia, with or
ficient of 0.9, meaning theoretical risks
without surgical
purposes of this review, mannitol and of it extravasating through the blood-
intervention depending
hypertonic saline will be discussed be- brain barrier into damaged brain tissue on the nature of the
cause they are the most commonly and exacerbating brain shift exist. How- intracranial pathology.
used and widely available agents. Other ever, this theoretical concern is not seen
h Mannitol can be given
medical interventions such as intuba- in practice, and mannitol remains the through a peripheral IV,
tion, sedation, mechanical ventilation, most commonly used agent for ICP which is useful in
and neuromuscular paralysis are typi- crisis. Mannitol is frequently chosen be- emergent scenarios,
cally reserved for refractory ICP cases cause it is widely available and easy to whereas 23.4%
after mannitol or hypertonic saline (ie, administer through a peripheral IV line.44 hypertonic saline
osmotherapy) has failed. An overview One concern regarding mannitol is solutions should be given
of therapies used to treat ICP is pro- overshooting hypernatremia from vol- through a central line.
vided in Table 3-3, and an algorithmic ume depletion and electrolyte deple-
approach is shown in Figure 3-10. tion (eg, total body sodium, potassium,
magnesium, calcium), given its osmotic
MANNITOL AND HYPERTONIC diuretic effect. In patients who experi-
SALINE: OSMOTHERAPY ence volume and electrolyte depletion,
Mannitol is a large molecule, similar to volume replacement with physiologic
starch, used as an IV osmotic diuretic. saline (0.9%) and electrolyte replacement
Mannitol causes an increase in serum with IV or enteral potassium guided by
osmolality and an osmotic gradient be- laboratory monitoring is advised. The
tween the serum and intracranial com- frequency of laboratory monitoring of
partment with subsequent removal of electrolytes and serum osmolarity (Osm)
brain water to reduce ICP. Mannitol depends on the dose and dose fre-
also produces an increase in cerebral quency of mannitol. For example, a
blood volume (CBV) and CBF as it passes large dose of mannitol (100 g IV) can
intracranially because it is an osmotic precipitate a massive diuresis of up
diuretic that attracts water molecules. to 1 L to 2 L in 1 to 2 hours in patients
Mannitol, also owing to its large starchlike with normal kidney function. The
molecular structure, is believed to have decision to check osmolarity depends
blood rheologic properties by de- on the medical provider and labora-
creasing blood viscosity. Some use the tory methods. For example, some labo-
osmolal gap to calculate the unmea- ratories measure serum osmolarity, while
sured solute in blood. Osmolal gap is others simply calculate it from serum
determined by the measured osmo- electrolytes via the equation45: serum
lality minus the calculated osmolality. osmolarity=(2Na+ )+(BUN/2.8)+
Calculated osmolality is performed by (glucose/18)+(ethanol/4.6).
inputting the patient’s laboratory values What is the upper safety limit for
(Na = sodium, K = potassium, BUN = mannitol administration? Most centers
blood urea nitrogen) into the serum use serum sodium above 155 mEq/L
osmolality equation equal to (2  or 160 mEq/L as the threshold to stop

Continuum (Minneap Minn) 2015;21(5):1299–1323 www.ContinuumJournal.com 1315

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

 Intracranial Pressure

hyperosmolar therapy such as mannitol, ample, if 25 g of IV piggyback manni-

or a serum osmolarity of greater than tol is given as needed for ICP elevation
320 mOsm/kg. However, this serum greater than 20 mm Hg, this order is
osmolarity cutoff of 320 mOsm/kg is quite specific and can be used along
likely relative, and serum osmolarity with electrolyte/osmolarity monitoring
may be tolerated up to 340 mOsm/kg.45 every 4 to 6 hours with replacements
The main precautions with mannitol as needed. In contrast, giving sched-
uled mannitol at 25 g IV piggyback
administration are: (1) electrolyte de-
every 6 hours to prevent ICP surges
pletion and inherent risks such as
above 20 mm Hg could be employed
hypokalemia-induced ECG changes,
prophylactically, for example, as a tem-
(2) volume depletion and subsequent
porizing strategy until neurosurgical
hypotension, which can be avoided decompression for refractory ICP. Con-
using physiologic saline (0.9%) reple- tinuing such scheduled mannitol with-
tion, and (3) risks associated with pa- out an ICP, imaging, or clinical end point
tients with severe congestive heart can be pointless and unnecessary. This
failure or end-stage renal disease. For is why protocolized care of such pa-
patients with severe congestive heart tients is advised by recent guidelines.40
failure, mannitol may theoretically cause
transient volume overload by adding HYPERTONIC SALINE
to circulating blood volume but should Hypertonic saline (sodium chloride)
later act as a diuretic, assuming normal comes in various concentrations, such
kidney function. Giving mannitol to as 1.5%, 3%, 10%, 14.6%, and 23.4%. Ty-
pically, concentrations of hypertonic
patients with end-stage renal disease
saline of 3% or greater should be given
on hemodialysis who are anuric poses
through a central line to prevent periph-
a risk because these patients cannot eral IV thrombophlebitis. The mecha-
excrete mannitol from their kidneys. nism of hypertonic saline is similar to
In such patients, mannitol is relatively mannitol because it creates an osmotic
contraindicated unless there is a sub- gradient between the serum and the in-
sequent way to dialyze the patient.46 tracranial compartment to reduce ICP.
The decision to use boluses as needed, In fact, equal osmolar doses of manni-
or to schedule doses of mannitol, is at tol and hypertonic saline are similarly
the discretion of the medical provider effective in reducing ICP.47,48 How-
ever, hypertonic saline is different be-
and depends on the goals of therapy
cause it is not a diuretic and increases
(eg, acute clinical changes or main-
total body sodium and chloride con-
tenance of ICP below 20 mm Hg). The
tent. Therefore, volume depletion is
risks for giving mannitol at fre- not a major adverse effect for hyper-
quent scheduled intervals include elec- tonic saline. Repeated boluses of hyper-
trolyte and volume depletion if these tonic saline (14.6% or 23.4%) can be
are not replaced before the next man- effective as long as there is a protocol
nitol dose. While protocols can be de- to monitor serum sodium49 because
vised to replace volume (eg, 100 cc such patients can become iatrogeni-
urine output replaced with 100 cc IV cally salt and volume overloaded (eg,
0.9% saline) and electrolyte replace- anasarca). Notably, mannitol can be al-
ment, the real question is the target ternated with hypertonic saline in such
(ie, administration rules and stopping patients to remove total body sodium
points) for such osmotherapy. For ex- and volume.

1316 www.ContinuumJournal.com October 2015

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

 KEY POINT
The decision of whether to use saline or mannitol, enteral water pushes h Hypotonic IV fluids
mannitol versus hypertonic saline de- or even an enteral water infusion can should be avoided in
pends on several factors. For exam- be used. In cases of severe hyperna- patients with brain
ple, patients with severe subarachnoid tremia (greater than 165 mEq/L), 0.45% injury with intracranial
hemorrhage and vasospasm requiring normal saline or 0.45% saline with pressure elevation
hypertensive, hypervolemic, and he- dextrose 5% water infusion could be because they increase
modilutional (HHH) therapy might ben- used as a last resort. However, IV in- cerebral edema and
efit from hypertonic saline infusions fusion of these hypotonic fluid solu- intracranial pressure.
or boluses to reduce ICP without caus- tions should be avoided as much as
ing volume depletion and hypoten- possible, and in the great majority of
sion. Hypertonic saline has a reflection cases, the hypernatremia can be suf-
coefficient of 1.0, meaning there is a ficiently controlled with the adminis-
lower theoretical risk of extravasation tration of enteral water infusions and/or
across the blood-brain barrier into the boluses. Lastly, patients who develop
brain with the concern for causing high urine output and hypernatremia
worsening mass effect and shift. Current regardless of mannitol or hypertonic
American Heart Association/American saline use should be monitored for de-
Stroke Association guidelines do not velopment of diabetes insipidus that
favor one agent over the other 33,35 for can be a complication of the underly-
treatment of ICP, leaving the decision at ing severe brain injury. The diagnosis
the discretion of the medical provider. of central diabetes insipidus can con-
Monitoring of hypertonic saline also found the use of hyperosmolar agents
requires frequent measurements of such as mannitol, which also cause diure-
electrolytes and osmolarity. It is im- sis. The diagnosis of diabetes insipidus
portant to watch for signs of iatro- is important since patients with severe
genic salt (and volume) overload and brain injuries who clinically progress
for the possible development of a hy- to brain death commonly develop cen-
pernatremic, hyperchloremic state, as tral diabetes insipidus with secondary
this can cause nonanion gap meta- massive diuresis and severe hyperna-
bolic acidosis from relative bicarbon- tremia (serum sodium typically in ex-
ate depletion. When there is sodium cess of 165 mEq/L to 170 mEq/L).
excess, bicarbonate is often wasted by
the kidneys. This state can be corrected HYPOTHERMIA
by enteral free water flushes, although Mild to moderate hypothermia (32-C
in patients receiving hypertonic saline, to 34-C [89.6-F to 93.2-F]) is effective
there is no true water deficit, but in reducing ICP43 regardless of the un-
rather an excess of sodium chloride. derlying type of brain injury. Hypother-
In patients treated with mannitol, free mia reduces cerebral metabolism and
water deficit calculations are more ap- CBF in a temperature-dependent fash-
propriate because these patients can ion to reduce ICP. However, hypother-
indeed become dehydrated. mia does not improve outcomes of
Finally, in patients with ICP eleva- patients with TBI despite reduction in
tion treated with either mannitol or hy- ICP values when used for 24 hours
pertonic saline, IV hypotonic fluids are compared to other medical or surgi-
not advised because hypotonic IV fluids cal therapies.25,28 The National Acute
will worsen brain edema and further Brain Injury Study: Hypothermia II
increase ICP. If hypernatremia exceeds (NABIS II) study 26 was a randomized
the safety thresholds previously de- controlled trial of patients with severe
scribed in patients receiving hypertonic TBI receiving either normothermia or
Continuum (Minneap Minn) 2015;21(5):1299–1323 www.ContinuumJournal.com 1317

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

 Intracranial Pressure

KEY POINT
h Mild hyperventilation hypothermia (35-C [95-F] initially, and 34 mm Hg) transiently, followed by
(PaCO2 32 mm Hg to 33-C [91.4-F] for 48 hours if patients a return to normocapnia (Pa CO 2 of
35 mm Hg) should be met secondary criteria, with subsequent 35 mm Hg to 40 mm Hg) as other med-
used emergently as a gradual rewarming). This trial did not ical or surgical interventions to reduce
temporary measure. demonstrate a difference in functional ICP are instituted.
Prolonged and aggressive outcome at 6 months between the two
hyperventilation leads groups. Only one study to date dem- CORTICOSTEROIDS
to worse outcomes in onstrated improved outcomes with Corticosteroids are not effective in
patients with traumatic prolonged hypothermia (up to 5 days) reducing cytotoxic or interstitial edema,
brain injury. compared to nonhypothermia in pa- and their administration has been asso-
tients with TBI.27 ciated with worse outcomes in patients
The importance of normothermia with cerebral infarcts and TBI-related
in patients with brain injury cannot be edema.52 Corticosteroids are tempo-
overstated because fever (temperature rarily effective, however, in reducing
greater than 38-C [100.4-F]) is associ- vasogenic edema from both primary
ated with worsened neurologic out- and metastatic brain tumors.53,54
comes in many different types of brain
injury.50 Cooling devices designed to SURGICAL MANAGEMENT
hold patients at so-called controlled OPTIONS FOR INTRACRANIAL
normothermia (37-C [98.6-F]) are be- PRESSURE CONTROL
ing studied to prevent secondary neu- Craniectomy (ie, removal of part of
rologic injury due to fever.51 the cranial bone) is typically only per-
formed for medically refractory ICP
HYPERVENTILATION cases and in operable cases. Surgery
Hyperventilation is sometimes used is particularly beneficial for certain mass
as a temporary intervention but should lesions, most notably extra-axial hema-
be considered a last-ditch effort. Hy- tomas and large (greater than 3 cm in
perventilation is considered a bridge diameter) posterior fossa hemorrhage
therapy and not a destination therapy with mass effect.33 Patients with TBI
for patients with refractory ICP. Hyper- can benefit from evacuation of resect-
ventilation can be used to reduce ICP able mass lesions (such as hemorrhagic
temporarily, along with other interven- contusions). Patients with massive
tions, but should not be used alone or hemispheric brain infarctions can have
in an aggressive or prolonged fashion. improved outcomes with decom-
A TBI study demonstrated that pro- pressive hemicraniectomy. Conven-
longed, aggressive hyperventilation tional craniotomy for evacuation of
(PaCO2 of approximately 25 mm Hg) deep intracerebral hemorrhage (ICH)
resulted in worse outcomes.39 Carbon within the basal ganglia is not benefi-
dioxide is a potent vasodilator by re- cial, as demonstrated by several random-
ducing the brain interstitial pH and, ized trials, the largest and most recent
consequently, increasing ICP. Con- being the Surgical Treatment for Intra-
versely, hyperventilation reduces ICP cerebral Hemorrhage (STICH) trial.55
but the vasoconstriction induced by In general, there are two types of
the hypocapnia can cause ischemia (al- craniectomy: decompressive hemicra-
though this risk should only occur in niectomy, in which bone is removed
areas with preserved vasoreactivity). In over one-half of the skull, and bifrontal
intubated patients with ICP crises, mild craniotomy, in which a bifrontal bone
hyperventilation can be used (PaCO2 flap is removed. Typically, decompres-
on arterial blood gas to 32 mm Hg to sive hemicraniectomy is performed for
1318 www.ContinuumJournal.com October 2015

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

 KEY POINT
patients with large middle cerebral minimize the risks of invasive brain h Ventriculostomy is still
artery infarcts and malignant edema. probe or ventriculostomy monitoring, considered the gold
Decompressive hemicraniectomy for such as bleeding and infection. Recently, standard for intracranial
malignant middle cerebral artery infarc- studies have investigated the value of pressure measurement
tion has been studied in several ran- measuring the optic nerve sheath dia- given its historical
domized controlled trials and a recent meter in patients with increased ICP context, and is the
meta-analysis.56Y58 Randomized trial by comparing it against standard inva- modality that most
data demonstrate a reduction in mor- other probes are
sive ICP monitoring. In a prospective
tality when compared to conservative validated against.
study, three optic nerve sheath diam-
therapy, but with varying levels of neu- eter measurements were made using
rologic morbidity depending on the un- optic ultrasound over each eye and were
derlying severity of ischemic infarction compared against ventriculostomy-
and the age of the patient.58 For more measured ICP in 65 patients.60 The
information, refer to the article ‘‘Man-
results demonstrated that ICP mea-
agement of Hemispheric Infarction and
surements greater than 20 mm Hg by
Ischemic Swelling’’ by Kevin N. Sheth,
the standard method correlated with an
MD, FAHA, FCCM, FNCS, FAAN, FANA,
optic nerve sheath diameter of 0.48 cm
in this issue of Continuum.
or more (sensitivity of 96% and spe-
The value of bifrontal craniotomy
in patients with TBI was studied in the cificity of 94%). Another recent study
Decompressive Craniectomy in Dif- also demonstrated similar correlations
fuse Traumatic Brain Injury (DECRA) between optic nerve sheath diameter
trial.59 This randomized controlled trial by CT measurement and ICP measure-
demonstrated that ICP improved in ment in patients with TBI.61 Given con-
the bifrontal craniotomy group com- cerns about inter-rater reliability and
pared to the conservative nonsurgical external validity outside of a controlled
group and was associated with a re- trial, further studies are warranted to
duced intensive care unit length of stay. validate these noninvasive methods
However, the surgical intervention was before they can be recommended for
not associated with improved func- clinical practice.
tional outcomes. It is important to note
CONCLUSION
that while these craniectomies can be
lifesaving interventions, the eventual Large middle cerebral artery ischemic
functional outcome will be determined strokes with malignant edema, ICH,
by the severity of the cerebral injury subarachnoid hemorrhage, and TBI are
(either by ischemia or trauma) and the common causes of intracranial hyper-
rehabilitation potential of the patient. tension. Providers should be aware of
Therefore, the patient’s family or proxy the signs and symptoms of elevated
should be carefully advised to consider ICP and how to manage, diagnose, and
the long-term implications of survival treat this condition. ICP is a reflection
with moderately severe to severe neu- of the intracranial pressure-volume
rologic disability before opting for these state. ICP elevation may be suspected
on clinical grounds but requires con-
surgical interventions.
firmation with an invasive monitoring
TRENDS: NONINVASIVE device. ICP (and CPP) monitoring are
INTRACRANIAL PRESSURE recommended as part of a protocol-
MEASUREMENT driven care in patients at risk for intra-
A noninvasive method for ICP mea- cranial hypertension based on clinical
surement would certainly be useful to and imaging features29 and may guide
Continuum (Minneap Minn) 2015;21(5):1299–1323 www.ContinuumJournal.com 1319

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

 Intracranial Pressure

medical and surgical interventions. The crease in ICP, and Valsalva maneuver
precise ICP threshold for treatment re- during segmental muscle strength test-
mains uncertain but is most commonly ing increases it.
regarded as greater than 20 mm Hg to links.lww.com/CONT/A152
25 mm Hg or that which compromises B 2015 American Academy of Neurology.
CPP below 60 mm Hg to 65 mm Hg. The
indication and method for ICP moni- Supplemental Digital Content 3-2
toring should be individualized to the Low compliance and high elastance
specific diagnosis (eg, subarachnoid intracranial pressure (ICP) waveform.
hemorrhage or TBI). A standard inser- The ICP waveform shown demonstrates
tion and maintenance protocol should a value of greater than 20 mm Hg and
be used for ICP monitoring devices is frankly triangular with a low compli-
to reduce the risk of infection and other ance/high elastance appearance. CSF
complications. ICP waveform interpreta- is drained from the external ventricular
tion is important with regard to states drain (EVD) system (line goes flat for
of intracranial compliance/elastance a while) and is later reopened period-
and potential treatments. Centers man- ically. By draining CSF, this essentially
aging patients with elevated ICP should changes the ICP waveform by moving
consider developing protocols for the down and left on the elastance curve.
administration of osmotherapy (man- Later, the ICP waveform returns after
nitol, hypertonic saline) and for man- the external ventricular drain is opened
agement of refractory intracranial and some P wave components are seen.
hypertension. Refractory ICP is a marker However, it is important to recognize
of severity of the primary neurologic that the ICP waveform still has an over-
illness and is associated with increased all noncompliant morphology indica-
mortality, but ICP should not be used tive of a persistent abnormal intracranial
in isolation as a predictor for pro- pressure-volume state.
gnostic or functional outcome. Newer links.lww.com/CONT/A153
methods of noninvasive ICP mea- B 2015 American Academy of Neurology.
surement, such as optic nerve sheath
diameter, are being studied in com-
REFERENCES
parison to standard invasive methods, 1. Agrawal A, Timothy J, Cincu R, et al.
but they are not ready for routine Bradycardia in neurosurgery.
clinical application. Clin Neurol Neurosurg 2008;110(4):
321Y327. doi:10.1016/j.clineuro.2008.
01.013.
VIDEO LEGENDS
2. Kim DJ, Czosnyka Z, Kasprowicz M, et al.
Supplemental Digital Content 3-1 Continuous monitoring of the Monro-Kellie
Mildly abnormal intracranial pressure doctrine: is it possible? J Neurotrauma
2012;29(7):1354Y1363. doi:10.1089/
(ICP) waveform. Mild, abnormal ICP
neu.2011.2018.
waveform during external ventricular
3. Srinivasan VM, O’Neill BR, Jho D, et al. The
drain (EVD) clamp trial with cluster-
history of external ventricular drainage.
ing of P1 to P3 waves. Respiratory J Neurosurg 2014;120(1):228Y236.
variations are noted as well as Valsalva doi:10.3171/2013.6.JNS121577.
maneuver. The patient has muscle re- 4. Wu OC, Manjila S, Malakooti N, Cohen AR.
sistance testing of his deltoid, which The remarkable medical lineage of the
causes the patient to perform the Monro family: contributions of Alexander
primus, secundus, and tertius. J Neurosurg
Valsalva maneuver and transiently in- 2012;116(6):1337Y1346. doi:10.3171/
crease ICP. Inspiration causes a de- 2012.2.JNS111366.

1320 www.ContinuumJournal.com October 2015

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

 5. Monro A. Observations on the structure intracranial pressure from cerebral blood
and function of the nervous system. flow velocity and arterial pressure. Sci Transl
Edinburgh, UK: Creech & Johnson, 1783. Med 2012;4(129):129ra44. doi:10.1126/
6. Rengachary SS, Ellenbogen RG, eds. scitranslmed.3003249.
Principles of neurosurgery. Edinburgh, 20. Czosnyka M, Kirkpatrick PJ, Pickard JD.
UK: Elsevier Mosby, 2005. Multimodal monitoring and assessment of
7. Lundberg N. Monitoring of intracranial cerebral haemodynamic reserve after severe
pressure. Proc R Soc Med 1972;65(1):19Y22. head injury. Cerebrovasc Brain Metab Rev
1996;8(4):273Y295.
8. Risberg J, Lundberg N, Ingvar DH. Regional
cerebral blood volume during acute transient 21. Komotar RJ, Schmidt JM, Starke RM, et al.
rises of the intracranial pressure (plateau Resuscitation and critical care of poor-grade
waves). J Neurosurg 1969;31(3):303Y310. subarachnoid hemorrhage. Neurosurgery
2009;64(3):397Y410. doi:10.1227/
9. Rose JC, Mayer SA. Optimizing blood 01.NEU.0000338946.42939.C7.
pressure in neurological emergencies.
Neurocrit Care 2004;1(3):287Y299. 22. Stuart RM, Claassen J, Schmidt M, et al.
Multimodality neuromonitoring and
10. Brain Trauma Foundation; American decompressive hemicraniectomy after
Association of Neurological Surgeons; subarachnoid hemorrhage. Neurocrit Care
Congress of Neurological Surgeons; Joint 2011;15(1):146Y150. doi:10.1007/
Section on Neurotrauma and Critical Care, s12028Y009Y9264Y2.
AANS/CNS, Bratton SL, Chestnut RM, Ghajar J,
et al. Guidelines for the management 23. Zweifel C, Dias C, Smielewski P, Czosnyka M.
of severe traumatic brain injury. VII. Continuous time-domain monitoring of
Intracranial pressure monitoring technology. cerebral autoregulation in neurocritical care.
J Neurotrauma 2007;24(suppl 1):S45YS54. Med Eng Phys 2014;36(5):638Y645.
doi:10.1016/j.medengphy.2014.03.002.
11. Chesnut RM, Temkin N, Carney N,
et al. A trial of intracranial-pressure 24. Reynolds SF, Heffner J. Airway management
monitoring in traumatic brain injury. of the critically ill patient: rapid-sequence
N Engl J Med 2012;367(26):2471Y2481. intubation. Chest 2005;127(4):1397Y1412.
doi:10.1056/NEJMoa1207363. doi:10.1378/chest.127.4.1397.
12. Chesnut RM, Marshall LF. Management 25. Polderman KH. Induced hypothermia and
of head injury. Treatment of abnormal fever control for prevention and treatment
intracranial pressure. Neurosurg of neurological injuries. Lancet 2008;
Clin N Am 1991;2(2):267Y284.
371(9628):1955Y1969. doi:10.1016/
13. Eide PK, Brean A. Cerebrospinal fluid pulse S0140Y6736(08)60837Y5.
pressure amplitude during lumbar infusion in
26. McCauley SR, Wilde EA, Moretti P, et al.
idiopathic normal pressure hydrocephalus can
Neurological outcome scale for traumatic
predict response to shunting. Cerebrospinal
Fluid Res 2010;7:5. doi:10.1186/1743Y8454Y7Y5. brain injury: III. Criterion-related validity
and sensitivity to change in the NABIS
14. Drummond JC. Elastance versus compliance. hypothermia-II clinical trial. J Neurotrauma 2013;
Anesthesiology 1995;82(5):1309Y1310. 30(17):1506Y1511. doi:10.1089/neu.2013.2925.
15. Cyrous A, O’Neal B, Freeman WD. New 27. Jiang JY, Xu W, Li WP, et al. Effect of long-term
approaches to bedside monitoring mild hypothermia or short-term mild hypothermia
in stroke. Expert Rev Neurother 2012;12(8): on outcome of patients with severe traumatic
915Y928. doi:10.1586/ern.12.85. brain injury. J Cereb Blood Flow Metab 2006;26(6):
16. Peace K, Wilensky EM, Frangos S, et al. The use 771Y776. doi:10.1038/sj.jcbfm.9600253.
of a portable head CT scanner in the intensive
28. Kramer C, Freeman WD, Larson JS,
care unit. J Neurosci Nurs 2010;42(2):109Y116.
et al. Therapeutic hypothermia for severe
doi:10.1097/JNN.0b013e3181ce5c5b.
traumatic brain injury: a critically appraised
17. Peace K, Maloney-Wilensky E, Frangos S, et al. topic. Neurologist 2012;18(3):173Y177.
Portable head CT scan and its effect on doi:10.1097/NRL.0b013e318253f8ef.
intracranial pressure, cerebral perfusion pressure,
and brain oxygen. J Neurosurg 2011;114(5): 29. Le Roux P, Menon DK, Citerio G, et al. Consensus
1479Y1484. doi:10.3171/2010.11.JNS091148. Summary Statement of the International
Multidisciplinary Consensus Conference on
18. Swallow DM, Fellner N, Varsos GV, et al. Multimodality Monitoring in Neurocritical
Repeatability of cerebrospinal fluid constant Care: a statement for healthcare professionals
rate infusion study. Acta Neurol Scand
from the Neurocritical Care Society and the
2014;130(2):131Y138. doi:10.1111/ane.12246.
European Society of Intensive Care Medicine.
19. Kashif FM, Verghese GC, Novak V, et al. Neurocrit Care 2014;21(suppl 2):1Y26.
Model-based noninvasive estimation of doi:10.1007/s12028Y014Y0041Y5.

Continuum (Minneap Minn) 2015;21(5):1299–1323 www.ContinuumJournal.com 1321

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

 Intracranial Pressure

30. Kochanek PM, Carney N, Adelson PD, 38. Brain Trauma Foundation; American Association
et al. Guidelines for the acute medical of Neurological Surgeons; Congress of
management of severe traumatic brain injury Neurological Surgeons. Guidelines for the
in infants, children, and adolescentsVsecond management of severe traumatic brain injury.
edition. Pediatr Crit Care Med 2012; J Neurotrauma 2007;24(suppl 1):S1YS106.
13(suppl 1):S1YS82. doi:10.1097/ 39. Muizelaar JP, Marmarou A, Ward JD,
PCC.0b013e31823f435c. et al. Adverse effects of prolonged
31. Carney NA, Chesnut R, Kochanek PM; hyperventilation in patients with severe
American Association for Surgery of Trauma; head injury: a randomized clinical trial.
Child Neurology Society; International Society J Neurosurg 1991;75(5):731Y739.
for Pediatric Neurosurgery; International 40. Stocchetti N, Picetti E, Berardino M,
Trauma Anesthesia and Critical Care Society; et al. Clinical applications of intracranial
Society of Critical Care Medicine; World pressure monitoring in traumatic brain
Federation of Pediatric Intensive and Critical injury: report of the Milan consensus
Care Societies. Guidelines for the acute conference. Acta Neurochir (Wien)
medical management of severe traumatic 2014;156(8):1615Y1622. doi:10.1007/
brain injury in infants, children, and adolescents. s00701Y014Y2127Y4.
Pediatr Crit Care Med 2003;4(3 suppl):S1.
41. Chesnut RM. Intracranial pressure. In:
32. Chesnut RM, Petroni G, Rondina C. Le Roux P, Levine J, Kofke P, eds. Monitoring
Intracranial-pressure monitoring in in neurocritical care. 2013:338Y347.
traumatic brain injury. N Engl J Med
2013;368(18):1751Y1752. doi:10.1056/ 42. Mack WJ, King RG, Ducruet AF,
NEJMc1301076. et al. Intracranial pressure following
aneurysmal subarachnoid hemorrhage:
33. Morgenstern LB, Hemphill JC 3rd, Anderson C, monitoring practices and outcome data.
et al. Guidelines for the management of Neurosurg Focus 2003;14(4):e3.
spontaneous intracerebral hemorrhage: a
guideline for healthcare professionals from the 43. Lozier AP, Sciacca RR, Romagnoli MF,
American Heart Association/American Stroke Connolly ES Jr. Ventriculostomy-related
Association. Stroke 2010;41(9):2108Y2129. infections: a critical review of the
doi:10.1161/STR.0b013e3181ec611b. literature. Neurosurgery 2008;
62(suppl 2):688Y700. doi:10.1227/
34. Bederson JB, Connolly ES Jr, Batjer HH, 01.neu.0000316273.35833.7c.
et al. Guidelines for the management of
aneurysmal subarachnoid hemorrhage: 44. Allen CH, Ward JD. An evidence-based approach
to management of increased intracranial
a statement for healthcare professionals
pressure. Crit Care Clin 1998;14(3):485Y495.
from a special writing group of the Stroke
Council, American Heart Association. 45. Diringer MN, Zazulia AR. Osmotic therapy:
Stroke 2009;40(3):994Y1025. doi:10.1161/ fact and fiction. Neurocrit Care 2004;
STROKEAHA.108.191395. 1(2):219Y233. doi:10.1385/NCC:1:2:219.

35. Connolly ES Jr, Rabinstein AA, Carhuapoma JR, 46. Yang B, Xu J, Xu F, et al. Intravascular
et al. Guidelines for the management of administration of mannitol for acute kidney
injury prevention: a systematic review and
aneurysmal subarachnoid hemorrhage:
meta-analysis. PloS One 2014;9(1):e85029.
a guideline for healthcare professionals
doi:10.1371/journal.pone.0085029.
from the American Heart Association/
American Stroke Association. Stroke 2012; 47. Francony G, Fauvage B, Falcon D, et al.
43(6):1711Y1737. doi:10.1161/ Equimolar doses of mannitol and hypertonic
STR.0b013e3182587839. saline in the treatment of increased intracranial
pressure. Crit Care Med 2008;36(3):795Y800.
36. Rabinstein AA, Atkinson JL, Wijdicks EF. doi:10.1097/CCM.0B013E3181643B41.
Emergency craniotomy in patients worsening
due to expanded cerebral hematoma: to what 48. Fink ME. Osmotherapy for intracranial
hypertension: mannitol versus hypertonic
purpose? Neurology 2002;58(9):1367Y1372.
saline. Continuum (Minneap Minn)
doi:10.1212/WNL.58.9.1367. 2012;18(3):640Y654. doi:10.1212/
37. Holloway RG, Arnold RM, Creutzfeldt CJ, 01.CON.0000415432.84147.1e.
et al. Palliative and end-of-life care in stroke: 49. Lewandowski-Belfer JJ, Patel AV, Darracott
a statement for healthcare professionals RM, et al. Safety and efficacy of repeated
from the American Heart Association/ doses of 14.6 or 23.4% hypertonic saline
American Stroke Association. Stroke for refractory intracranial hypertension.
2014;45(6):1887Y1916. doi:10.1161/STR. Neurocrit Care 2014;20(3):436Y442.
0000000000000015. doi:10.1007/s12028Y013Y9907Y1.

1322 www.ContinuumJournal.com October 2015

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

50. Kilpatrick MM, Lowry DW, Firlik AD, et al. Neurol 2003;60(3):227Y232. doi:10.1016/
Hyperthermia in the neurosurgical intensive S0090Y3019(03)00266Y0.
care unit. Neurosurgery 2000;47(4):850Y855.
57. Juttler E, Schwab S, Schmiedek P, et al.
51. Marion DW. Controlled normothermia in Decompressive Surgery for the Treatment
neurologic intensive care. Crit Care Med of Malignant Infarction of the Middle
2004;32(2 suppl):S43YS45. doi:10.1097/ Cerebral Artery (DESTINY): a randomized,
01.CCM.0000110731.69637.16. controlled trial. Stroke 2007;38(9):
2518Y2525. doi:10.1161/
52. Roberts I, Yates D, Sandercock P, et al. Effect
STROKEAHA.107.485649.
of intravenous corticosteroids on death within
14 days in 10008 adults with clinically 58. Vahedi K, Hofmeijer J, Juettler E, et al. Early
significant head injury (MRC CRASH trial): decompressive surgery in malignant infarction
randomised placebo-controlled trial. Lancet 2004; of the middle cerebral artery: a pooled
364(9442):1321Y1328. doi:10.1186/cc3813. analysis of three randomised controlled trials.
Lancet Neurol 2007;6(3):215Y222. doi:10.
53. Ryken TC, McDermott M, Robinson PD, 1016/S1474Y4422(07)70036Y4.
et al. The role of steroids in the management
of brain metastases: a systematic review 59. Cooper DJ, Rosenfeld JV, Murray L, et al.
and evidence-based clinical practice guideline. Decompressive craniectomy in diffuse
J Neurooncol 2010;96(1):103Y114. traumatic brain injury. N Engl J Med
doi:10.1007/s11060Y009Y0057Y4. 2011;364(16):1493Y1502. doi:10.1056/
NEJMoa1102077.
54. Bebawy JF. Perioperative steroids for peritumoral
intracranial edema: a review of mechanisms, 60. Rajajee V, Vanaman M, Fletcher JJ, Jacobs
efficacy, and side effects. J Neurosurg TL. Optic nerve ultrasound for the detection
Anesthesiol 2012;24(3):173Y177. doi:10.1097/ of raised intracranial pressure. Neurocrit
ANA.0b013e3182578bb5. Care 2011;15(3):506Y515. doi:10.1007/
s12028Y011Y9606Y8.
55. Morgenstern LB, Frankowski RF, Shedden P,
et al. Surgical treatment for intracerebral 61. Sekhon MS, Griesdale DE, Robba C,
hemorrhage (STICH): a single-center, et al. Optic nerve sheath diameter on
computed tomography is correlated with
randomized clinical trial. Neurology 1998;
simultaneously measured intracranial
51(5):1359Y1363. doi:10.1212/WNL.51.5.1359.
pressure in patients with severe
56. Cho DY, Chen TC, Lee HC. Ultra-early traumatic brain injury. Intensive Care Med
decompressive craniectomy for malignant 2014;40(9):1267Y1274. doi:10.1007/
middle cerebral artery infarction. Surg s00134Y014Y3392Y7.

Continuum (Minneap Minn) 2015;21(5):1299–1323 www.ContinuumJournal.com 1323

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.