CT Scan Head N Brain

CHAPTER 1 Imaging the Head and Brain
Joshua Broder, MD, FACEP, and Robert Preston, MD
Emergency physicians frequently evaluate patients detail in Chapter 8 in the context of cardiac imaging)
with complaints requiring brain imaging for diagnosis have dramatically enhanced the resolution and diag-
and treatment. The diversity of imaging modalities and nostic utility of CT since its introduction. CT relies on
variations of these modalities may be daunting, creating the differential attenuation of x-ray by body tissues of
uncertainty about the most appropriate, sensitive, and differing density. The image acquisition occurs by rapid
specific modality to evaluate the presenting complaint. movement of the patient through a circular gantry
An evidence-based approach is essential, with modality opening equipped with an x-ray source and multiple
and technique chosen based on patient characteristics detectors. A three-dimensional volume of image data is
and differential diagnosis. In this chapter, we begin with acquired; this volume can be displayed as axial, sagit-
a brief summary of computed tomography (CT) and tal, or coronal planar slices or as a three-dimensional
magnetic resonance (MR) technology. Next, we present a image. CT does raise some safety concerns with regard
systematic approach to interpretation of head CT, along to long-term biologic effects of ionizing radiation and
with evidence for interpretation by emergency physi- carcinogenesis, which we describe later. The radiation
cians. Then, we discuss the cost and radiation exposure exposure to the fetus in a pregnant patient undergo-
from neuroimaging, as these are important reasons to ing head CT is minimal.3 Most commercially available
limit imaging. We review the evidence supporting the CT scanners have a weight capacity of approximately
use of CT and magnetic resonance imaging (MRI) for 450 pounds (200 kg), although some manufactur-
diagnosis and treatment of emergency brain disorders, ers now offer units with capacities up to 660 pounds
concentrating on clinical decision rules to target imag- (300 kg).4 Portable dedicated head CT scanners with
ing to high-risk patients. We also consider adjunctive acceptable diagnostic quality and no weight limits are
imaging techniques, including conventional angiogra- now available.4a
phy, plain films, and ultrasound. By chapter’s end, we
consider the role of neuroimaging in the evaluation of Noncontrast Head CT
headache, transient ischemic attacks (TIAs) and stroke, Noncontrast CT is the most commonly ordered head
seizure, syncope, subarachnoid hemorrhage (SAH), imaging test in the ED, used in up to 12% of all adult
meningitis, hydrocephalus and shunt malfunction, and ED visits.5,6 It provides information about hemorrhage,
head trauma. ischemic infarction, masses and mass effect, ventricular
abnormalities such as hydrocephalus, cerebral edema,
sinus abnormalities, and bone abnormalities such as
NEUROIMAGING MODALITIES fractures. Although dedicated facial CT provides more
Indications for neuroimaging are diverse, including detail by acquiring thinner slices through the region of
traumatic and nontraumatic conditions (Table 1-1). The interest or by changing the patient’s position in the scan-
major brain neuroimaging modalities today are CT and ner during image acquisition, general information about
MRI, with adjunctive roles for conventional angiogra- the face and sinuses can be gleaned from a generic non-
phy and ultrasound. Plain films of the calvarium have contrast head CT, as described in detail in Chapter 4 on
an extremely limited role, as they can detect bony injury facial imaging. The American College of Radiology rec-
but cannot detect underlying brain injury, which may be ommends acquisition of contiguous or overlapping slices
present even in the absence of fracture. with slice thickness no greater than 5mm. For imaging of
the cranial base, the ACR recommends the thinnest slices
Computed Tomography possible. If multiplanar or three-dimensional reconstruc-
CT has been in general clinical use in emergency depart- tion is planned, slice thickness should be as thin as pos-
ments (EDs) in the United States since the early 1980s. sible and no greater than 2-3 mm.6a
The modality was simultaneously and independently
developed by the British physicist Godfrey N. Houn- Contrast-Enhanced Head Computed Tomography
sfield and the American Allan M. Cormack in 1973, and Contrast-enhanced head CT is usually performed follow-
the two were corecipients of the Nobel Prize for Medi- ing a noncontrast CT, and the two are then compared. In
cine in 1979.1,2 Advances in computers and the intro- a contrast-enhanced head CT, IV contrast is injected, usu-
duction of multislice helical technology (described in ally through an upper extremity intravenous catheter.
1
2 DIAGNOSTIC IMAGING FOR THE EMERGENCY PHYSICIAN
TABLE 1-1. Clinical Indications, Differential Diagnoses, and Initial Imaging Modality
Clinical Indication Differential Diagnosis Initial Imaging Modality
Headache Mass, traumatic or spontaneous hemorrhage, Noncontrast CT
meningitis, brain abscess, sinusitis, hydrocephalus
Altered mental status Mass, traumatic or spontaneous hemorrhage, Noncontrast CT
or coma meningitis, brain abscess, hydrocephalus
Fever Meningitis (assessment of ICP), brain abscess Noncontrast CT
Focal neurologic deficit— Mass, ischemic infarct, traumatic or spontaneous Noncontrast CT, possibly followed by
motor, sensory, hemorrhage, meningitis, brain abscess, sinusitis, MRI, MRA, or CTA, depending on
or language deficit hydrocephalus context
Focal neurologic com- Posterior fossa or brainstem abnormalities, vascular MRI or MRA of brain and neck; CT
plaint—ataxia or cranial dissections or CTA of brain and neck if MR is not
nerve abnormalities rapidly available
Seizure Mass, traumatic or spontaneous hemorrhage, Noncontrast CT, possibly followed by CT
meningitis, brain abscess, sinusitis, hydrocephalus with IV contrast or MR
Syncope Trauma Little indication for imaging for cause of
syncope, only for resulting trauma
Trauma Hemorrhage, mass effect, cerebral edema Noncontrast CT—if clinical decision rules
suggest need for any imaging
Traumatic loss of Hemorrhage, DAI, mass effect, cerebral edema Little indication when transient loss of
consciousness consciousness is isolated complaint
Planned LP Increased ICP Noncontrast head CT—limited indications
A time delay is introduced to allow the venous contrast field. MR provides outstanding soft-tissue contrast that
to pass to the brain. Depending on the delay between is superior to that obtained from CT. MR can be per-
contrast injection and image acquisition, the result- formed with or without intravenous contrast agents.
ing CT may be a CT cerebral angiogram (CTA) or a CT Advantages of MRI include the noninvasive nature of
cerebral venogram (CTV). The CT data may be recon- the test and its apparent safety in pregnancy.8 It does
structed in any of several planes (typically axial, sagittal, not employ ionizing radiation and has no known per-
or coronal) or even three-dimensionally. Cerebral perfu- manent harmful biologic effects.9 Traditionally, contra-
sion can be mapped with IV contrast, using a technique indications have been thought to include the presence
described in more detail later. Contrast CT is useful for of ferromagnetic material within the body, including
depicting abnormal vascular structures, such as aneu- electronic devices such as pacemakers or metallic debris
rysms or arteriovenous malformations; for demonstrat- such as shrapnel, especially in sensitive structures such
ing abnormal failure of filling of vascular structures, as the eye or brain. However, there are now more than
such as sagittal sinus thrombosis (akin to demonstrating 230 published prospective cases of patients with pace-
a filling defect in chest CT for pulmonary embolism); makers safely having undergone low-field MRI, so MRI
and for demonstrating neoplastic, inflammatory, and may be an imaging option in these patients.10 Magnetic
infectious processes. These latter abnormalities typi- effects on tattoos, including first-degree burns and
cally have increased regional blood flow, consequently burning sensation, have been reported, although these
receive abnormally high levels of vascular contrast appear rare and more likely to interfere with completion
agents, and may demonstrate ring enhancement, a cir- of MR than to cause significant harm.11–13
cumferential increase in density around a lesion on CT
occurring after IV contrast administration.
Can Emergency Physicians Accurately Interpret
Magnetic Resonance Head CT Images? What Are the Potential Benefits
MR has been in wide clinical use in the United States to Patients?
since the late 1980s. The modality was coinvented by Head CT is rapid to obtain, but delays in interpreta-
the American Paul C. Lauterbur and the British physi- tion could result in adverse patient outcomes if clini-
cist Sir Peter Mansfield, who shared the 2003 Nobel cal treatment decisions cannot be made in a timely
Prize in Medicine for their work.7 MR allows imaging fashion. Surveys of emergency medicine residency
of the brain by creating variations in the gradient of programs suggest that, in many cases, radiology inter-
a magnetic field and analyzing the radio waves emit- pretation is not rapidly available for clinical decisions
ted in response by hydrogen ions (protons) within the and that emergency physicians often perform the initial
IMAGING THE HEAD AND BRAIN 3
interpretation of radiographic studies. A study simulat- by neuroradiologists.17,18 Undoubtedly, improvements

ing a teleradiology support system estimated the time in training are needed, but the pragmatic limitations on
to interpretation of a noncontrast head CT at 39 min- the availability of subspecialist radiologists, even with
utes, potentially wasting precious time in patients with teleradiology, mean that emergency physicians must
intracranial hemorrhage or ischemic stroke.14 The ability become proficient first-line readers of emergency CT.
of the on-scene emergency physician to interpret the CT
could be extremely valuable. INTERPRETATION OF NONCONTRAST
HEAD CT
Multiple studies have examined the ability of emer-
gency medicine residents and attending physicians Several systematic methods for interpretation of non-
to interpret head CT. A 1995 study showed that in an contrast head CT have been described. The mnemonic
emergency medicine residency program, although up “Blood Can Be Very Bad” has been shown to assist in the
to 24% of potentially significant CT abnormalities were sustained improvement of interpretation by emergency
not identified by the emergency medicine residents, medicine residents.16 The mnemonic reminds the inter-
only 0.6% of patients appear to have been misman- preter to look for blood (blood), abnormalities of cisterns
aged as a result.15 Studies have shown that substantial and ventricles (can and very, respectively), abnormalities
and sustained improvements in interpretation ability of the brain parenchyma (be), and fractures of bone (bad).
can occur with brief training. Perron et al.16 showed
an immediate improvement from 60% to 78% accuracy Broder used the familiar ABC paradigm to drive the
after a 2-hour training session based on a mnemonic, assessment of the head CT (see “A Mnemonic for Head
sustained at 3 months. In the setting of stroke, emer- CT Interpretation: ABBBC”).5
gency medicine attending physicians perform rela-
tively poorly in the recognition of both hemorrhage and Hounsfield Units and Windows
early ischemic changes, which may contraindicate tis- The density of a tissue is represented using the Houn-
sue plasminogen activator (t-PA) administration, with sfield scale, with water having a value of zero Houn-
accuracy of approximately 60%. However, neurologists sfield units (HU), tissues denser than water having
and general radiologists achieve only about 80% accu- positive values, and tissues less dense than water having
racy compared with the gold-standard interpretation negative values (Figure 1-1). By convention, low-density
bone (≈+1000HU) bone (≈+1000HU)

A B
brain brain
(≈+40HU) (≈+40HU)
subdural subdural
hematoma hematoma
(≈+60HU) (≈+60HU)
air outside air outside

of patient of patient
(-1000HU) (-1000HU)
CSF (water) CSF (water)

subcutaneous (0HU) subcutaneous (0HU)
fat (-50HU) fat (-50HU)
Figure 1-1. The CT Hounsfield scale and window settings. The CT Hounsfield scale places water density at a value of zero with air and bone at
opposite extreme values of -1000HU and +1000HU. Fat is less dense than water and has a density around -50HU. Other soft tissues are slightly more
dense than water and have densities ranging from around +20 to +100HU. The colors associated with these density values can be reassigned to highlight
particular tissues, a process called “windowing.” A, Axial CT slice, viewed with brain window settings. Notice in the grayscale bar at the right side of the
figure that the full range of shades from black to white has been distributed over a narrow HU range, from zero (pure black) to +100HU (pure white).
This allows fine discrimination of tissues within this density range, but at the expense of evaluation of tissues outside of this range. A large subdural
hematoma is easily discriminated from normal brain, even though the two tissues differ in density by less than 100HU. Any tissues greater than +100HU
in density will appear pure white, even if their densities are dramatically different. Consequently, the internal structure of bone cannot be seen with this
window setting. Fat (-50HU) and air (-1000HU) cannot be distinguished with this setting, as both have densities less than zero HU and are pure black.
B, The same axial CT slice viewed with a bone window setting. Now the scale bar at the right side of the figure shows the grayscale to be distributed over
a very wide HU range, from -450HU (pure black) to +1050HU (pure white). Air can easily be discriminated from soft tissues on this setting because it
is assigned pure black, while soft tissues are dark gray. Details of bone can be seen, because a large portion of the total range of gray shades is devoted
to densities in the range of bone. Soft tissue detail is lost in this window setting, because the range of soft tissue densities (-50HU to around +100HU)
represents a narrow portion of the gray scale.
tissues are assigned darker (blacker) colors and high- appear bright white, and internal detail of these high-
density structures are assigned brighter (whiter) col- density structures is lost.
ors. Because the human eye can perceive only a limited
number of gray shades, the full range of density values Right–Left Orientation of Computed Tomography
is typically not displayed for a given image. Instead, the Images
tissues of interest are highlighted by devoting the vis- By convention, axial CT images are displayed with the
ible gray shades to a narrow portion of the full density patient’s right side on the left side of the video screen or
range, a process called “windowing” (Figures 1-1 and printed image. The point of view is as follows. Imagine
1-2). The same image data can be displayed in different yourself standing at the foot of the CT gantry, gazing
window settings to allow evaluation of injury to differ- toward the patient’s head, with the patient positioned
ent tissues. In general, head CT images are viewed on supine. Your left hand is placed on the patient’s right
brain or bone windows to allow most emergency pathol- foot. The patient is virtually “sectioned” into slices like
ogy to be assessed (see Figure 1-2). a loaf of bread (Figure 1-3).
Bone windows are useful for evaluation in the setting of right left right
trauma. By shifting the gray scale to center on the range
of densities typical of bone bone windows allow detec-
tion of abnormalities such as subtle fracture lines. At the
same time, they sacrifice detailed evaluation of struc-
tures less dense than bone (brain, cerebrospinal fluid,
and blood vessels). Air remains black on bone windows
and can be readily identified—for example, intracranial
air can easily be seen on bone window settings. Sinus
spaces are also nicely delineated on bone windows
because of the contrast between air (black) and all other
tissues (gray shades).
left
Brain windows are useful for evaluation of brain hem-
Figure 1-3. Right–left orientation in computed tomography
orrhage, fluid-filled structures including blood vessels (CT). Axial CT images are displayed with the patient’s right side on the
and ventricles, and air-filled spaces. The majority of our left side of the video screen or printed image (A). The point of view is that
evaluations will be done using this window setting. On of an observer standing at the feet of a supine patient, gazing toward the
patient’s head. The volume of the patient’s head is divided into a stack of
brain windows, bone and other dense or calcified struc- axial slices for review. (B). (From Broder J. Midnight radiology: Emergency
tures (e.g., surgical clips and calcified pineal glands) all CT of the head. (Accessed at http://emedhome.com/.)
fracture
epidural
hematoma
intraventricular
hemorrhage
A B
Figure 1-2. Hounsfield units and windows. A single noncontrast computed tomography slice, shown in brain windows (A) and bone windows
(B). Brain windows allow evaluation of the brain parenchyma, hemorrhage, cerebrospinal fluid (CSF) spaces, and other soft tissue at the expense of
bony detail. Gross fractures may be seen. Bone windows allow detailed examination for fractures but obscure most soft-tissue detail. (From Broder J,
Preston R: An evidence-based approach to imaging of acute neurological conditions. Emerg Med Pract 9(12):4, 2007.)
midline
Brain Symmetry
The brain, air and cerebrospinal fluid (CSF) spaces and
the surrounding bone are normally symmetrical struc-
tures (Figure 1-4; see also Figures 1-7 and 1-19). A head
CT should be inspected for normal symmetry, as devia-
tion from this norm often indicates pathology (Figure
1-5; see also Figure 1-4). If the patient’s head is not
centered symmetrically in the CT gantry, the resulting
images can create a false sense of asymmetry. It is criti-
cal to note that some life-threatening pathological condi-
tions such as diffuse subarachnoid hemorrhage, cerebral
edema, and hydrocephalus can have a symmetrical CT
appearance, so symmetry does not guarantee normalcy.
Abnormal Asymmetry: Mass Effect

and Midline Shift
The compression or displacement of normal brain struc-
tures (including ventricles and sulci) by adjacent masses
is called mass effect. This displacement may occur due
to tumor, hemorrhage, edema, or obstruction of CSF
flow, to name but a few common causes. When this
effect becomes extreme, shift of brain structures across
the midline of the skull can occur, a finding called mid-
Figure 1-4. Normal brain symmetry brain windows. Abnormal line shift. Midline shift can indicate significant pathol-
masses or hemorrhage may deviate brain structures across an imaginary ogy, including threatened subfalcine herniation (Figure
line dividing the brain, creating midline shift. The degree of midline 1-5), and it should be carefully sought, as it may be more
shift is more important acutely than the exact etiology of the shift,
since shift is an indication of threatened subfalcine herniation and may important than the underlying etiology of the shift in
require surgical intervention. (From Broder J, Preston R: An evidence-based determining initial management. The degree of dis-
approach to imaging of acute neurological conditions. Emerg Med Pract placement of structures across the normal midline of the
9(12):6, 2007.)
brain can be easily measured using digital tools on the
midline
lateral
ventricles
lateral
ventricles
A B
Figure 1-5. Mass effect and midline shift, brain windows. Abnormal masses such as hemorrhage create mass effect, displacing other
structures from their normal positions. When mass effect drives structures across the brain midline, midline shift is present. The degree of mass
effect or midline shift is more important than the underlying etiology. Guidelines call significant midline shift (5 mm or greater) a neurosurgical
emergency, although the clinical status of the patient is the more important parameter. A, A large subdural hematoma (arrowheads) creates
a mass effect and is displacing brain structures, including the lateral ventricles (arrowheads), across the midline (dotted line) to the patient’s right.
B, A large epidural hematoma (arrowheads) creates mass effect. The lateral ventricles (arrowheads) are shifted across the midline (dotted line) to the
patient’s left.
picture archiving and communication system (PACS), a imperfect scan. Modern CT scanners acquire images at
computer viewing system. Midline shift may have some a very fast rate—a 64-slice CT can scan the entire brain
prognostic value in determining the likelihood of regain- in approximately 5 seconds.22 As a consequence, CT is
ing consciousness after surgical decompression; patients less subject to motion artifact than in the past, although
with significant shift, greater than 10 mm, are more likely significant patient motion may still render images unin-
to benefit from decompression than are those with lesser terpretable. Just as in standard photographs, motion
degrees of shift.19 Patients with shift of 5 mm or greater results in a blurry CT image.
are more likely to have neurologic deficits requiring
long-term supervision than are those with lesser midline Very dense objects create distortion on CT, called streak
shift.20 Midline shift is also linked to probability of death artifact. Examples include implanted metallic devices,
after traumatic brain injury.21 Published guidelines on such as cochlear implants and dental fillings; metallic for-
surgical indications for brain lesions include midline eign bodies, such as bullets; and even dense bone, such
shift as one of several parameters (as shown later), so as occipital bone surrounding the posterior fossa. These
recognizing and measuring midline shift is important. artifacts may make it difficult or impossible to identify
pathologic changes in the region (see Figure 1-6).
Artifacts: Motion and Metal
A brain CT should be examined for artifacts that may A MNEMONIC FOR HEAD CT
limit interpretation, including motion and streak arti-
INTERPRETATION: ABBBC
fact or beam-hardening artifact from high-density struc-
tures such as metal (Figure 1-6). Although artifact may A systematic approach to interpretation of head CT is
degrade the overall quality of the study, useful diag- necessary to avoid missing important abnormalities. We
nostic information can often still be gleaned from an review one approach, with a discussion of the normal
dense object
creates
dense streak artifact
dense
object
object
outside of
outside of
patient
patient
creates
creates
streak
streak
artifact
artifact
streak
artifact
A B
C
Figure 1-6. Streak artifacts. Streak artifacts may result from extremely high-density structures, such as dental fillings, implanted devices, or metal
objects outside of the patient. A, Streak artifacts are most evident on brain windows. B, It is evident that the source of the artifact is very high-density
material on bone windows. Notice how the external objects are denser (whiter) than the patient’s calvarium. C, A cochlear implant (in another patient)
also creates significant streak artifact. Although streak artifact may obscure some important details, relevant clinical information may be obtained
from an imperfect scan. For example, in these scans, no midline shift is visible. (From Broder J, Preston R: An evidence-based approach to imaging of acute
neurological conditions. Emerg Med Pract 9(12):6, 2007.)
appearance of the brain. Although a detailed under- air space may indicate bleeding into that space, raising
standing of neuroanatomy will improve your head CT suspicion of a fracture of the surrounding bone (Figure
interpretation, our mnemonic avoids significant ana- 1-8). In the absence of trauma, opacification may indi-
tomic detail, as many clinical decisions don’t require cate sinus infection or inflammation, although this is a
this level of sophistication. Table 1-2 gives an overview nonspecific finding that may require no treatment in the
of the mnemonic, with images illustrating each finding absence of symptoms.
in the figures that follow.
Normal air spaces appear black on either brain or bone
A Is for Air Spaces windows, because air has the lowest Hounsfield den-
Our mnemonic starts with A, for air-filled spaces in the sity, negative 1000 HU. The frontal, maxillary, ethmoid,
head. The normal head contains air-filled spaces: the and sphenoid sinuses are normally air-filled, with no
frontal, maxillary, ethmoid, and sphenoid sinuses and thickening of mucosa or air–fluid levels. The mastoid
the mastoid air cells (Figure 1-7). Opacification of an air processes are normally spongy bone filled with tiny
space may occur because of fluids such as pus, mucus, pockets of air, the mastoid air cells. If these air spaces
or blood; mucosal edema; or because of tumor invasion become partially or totally opacified with fluid, this is
of the space. In the setting of trauma, opacification of an easily recognized as a gray shade. Dense bone surround-
ing the small mastoid air cells creates a “bloom artifact”
that can obscure the actual air spaces when viewed on
TABLE 1-2. A Mnemonic for Systematic Interpretation brain windows, making assessment for abnormal fluid
of Non-Contrast Head CT: ABBBC more difficult. This is minimized by viewing this region
using bone windows. Abnormal fluid appears gray on
Assess These For Signs of This Pathology
this setting as well. The relatively small ethmoid air cells
Structures…
are also best-assessed for fluid using bone windows. For
Air-filled spaces Fractures larger air-spaces, bloom artifact is minimal and brain or
Mastoid air-cells Infections
bone windows will usually reveal fluid. Recognizing
Sinuses
abnormalities of normally air-filled structures requires
Bones Fractures some basic knowledge of their normal location and con-
Blood Epidural hemorrhage figuration (see Figure 1-7). Box 1-1 summarizes abnor-
Intraparenchymal hemorrhage malities of sinus air spaces, which are discussed in more
Subarachnoid hemorrhage detail later.
Subdural hemorrhage
Brain Midline shift or mass effect Sinus Trauma
Edema Facial fractures are discussed in more detail in the dedi-
Infarction cated chapter on facial imaging (Chapter 2). In trauma,
Masses fractures through the bony walls of sinuses result in
CSF spaces Atrophy bleeding into the sinus cavity. While the trauma patient
Cisterns Edema remains in a supine position, this blood accumulates
Sulci Hydrocephalus in the dependent portion of the sinus, forming an air–
Ventricles Subarachnoid hemorrhage in these fluid level visible on CT. Previously existing sinus dis-
structures ease may be visible as circumferential sinus mucosal
ethmoid air frontal

maxillary sphenoid sinus
cells
sinus sinus
mastoid mastoid
air cells air cells
A B C
Figure 1-7. Air-filled spaces. The normal location and appearance of air-filled spaces when viewed on brain windows. Air-filled spaces are normally
black on both brain and bone windows. Even more detail of fine bony partitions can be seen by selecting bone windows. A through C move progressively
from caudad to cephalad. The structures are labeled on one side. Try to identify the same structure on the opposite side. (From Broder J, Preston R: An
evidence-based approach to imaging of acute neurological conditions. Emerg Med Pract 9(12):5, 2007.)
right maxillary left maxillary

sinus air sinus fluid
Box 1-1: CT Findings of: Sinus
and Mastoid Pathology
maxillary
sinus ll Window setting: Bone windows helpful for
fluid
smaller air-space evaluation.
right ll CT Appearance: Normally air filled (black),
zygomatic no air–fluid levels
arch ll Significance
fracture mm Air–fluid levels or opacification in the set-
ting of trauma may indicate fracture.
mm In the absence of trauma, air–fluid levels or
mucosal thickening may be too sensitive
and should not necessarily be equated to
bacterial sinusitis in the absence of strong
clinical evidence.
mm Mastoid opacification without trauma may
indicate mastoiditis when clinical signs and
symptoms are present.
mastoid
air cells
sinus air–fluid levels are present and disruption of the
posterior plate is suspected. When the mastoid air cells
are obliterated or opacified, suspect temporal bone frac-
ture. The normal side is a useful comparison.
Figure 1-8. Fluid in normally air-filled spaces. Fractures of the bony
walls of normally air-filled structures, including the sinuses and mastoid
air cells, can result in hemorrhage. Blood pooling in the dependent portion Sinus Infections
of these air-filled chambers results in an air–fluid level or sometimes in In the absence of trauma, sinus mucosal thickening and
complete opacification. This may be the only evidence of fracture. If no air–fluid levels may be normal findings. They should not
history of trauma is present, fluid in an air space may indicate infection,
such as sinusitis or mastoiditis. Viewed here on bone windows, the be used to make a diagnosis of bacterial sinusitis in the
patient’s right maxillary sinus has an air (black)–fluid (gray) level. The absence of strong clinical evidence, as they are nonspe-
left maxillary sinus is completely opacified. The fractures of the sinuses cific and may occur in allergic sinusitis or even asymp-
themselves are hard to identify, though a right zygomatic arch fracture
is visible. The mastoid air cells are normal and have no fluid. Following tomatic patients. The mastoid air cells are not normally
trauma, opacification of the mastoid air cells is a sign of temporal bone fluid filled, and in the presence of mastoid tenderness
fracture. (From Broder J, Preston R: An evidence-based approach to imaging of and erythema, their opacification on CT is evidence of
acute neurological conditions. Emerg Med Pract 9(12):6, 2007.)
mastoiditis (Figure 1-10). Sinus infections are discussed
in more detail in Chapter 2, Imaging the Face.
thickening, rather than as an air–fluid level. Inspect the
sinuses carefully for air–fluid levels, as these may indi- B Is for Bone
cate occult fractures. In trauma, opacification of sinuses The first B in our mnemonic is for bone. Following
should be considered evidence of fracture until proven trauma, bony fractures should be suspected, although
otherwise, as the fracture itself may be hard to identify they are often of less clinical significance than any
(Figures 1-8 and 1-9). The ethmoid sinuses are small underlying brain injury. Abnormalities of bone includ-
and may be completely opacified by blood in the event ing acute fractures are best identified on bone windows.
of fracture. Opacified ethmoid sinuses should increase Defects in the cortex of bone may indicate fracture, but
suspicion of a medial orbital blowout fracture. Air–fluid these must be distinguished from normal suture lines
levels in the maxillary sinus may be associated with (Figure 1-11). Comparing the contralateral side to the
inferior orbital blowout fractures, because the inferior side in question may help to distinguish fractures from
wall of the orbit is the superior wall of the maxillary normal sutures. Air–fluid levels within sinuses follow-
sinus. The frontal sinus is less easily fractured, as its ing trauma are likely blood resulting from fracture (see
anterior and posterior plates are thick and resistant to Figures 1-8 and 1-9). Pneumocephalus (air within the
trauma. Fracture of the anterior wall of the frontal sinus calvarium) may also be present and may provide an
is relatively less concerning, requiring plastic surgery additional clue to fractures communicating with sinus
or otolaryngology consultation. Fracture of the poste- spaces or the outside world (Figures 1-12 and 1-13). Air
rior wall of the frontal sinus is a potential neurosurgi- may take the form of large amorphous collections abut-
cal emergency due to communication of the sinus space ting the calvarium or small black spheres within hemor-
with the CSF. Look for intracranial air whenever frontal rhage associated with the fracture.
right zygomatic air-fluid levels in

arch fracture ethmoid sinuses
fracture soft-
tissue
air
fracture
fracture
air-fluid
level in
sphenoid
sinus
normal suture
soft-
tissue
fracture air
A B
Figure 1-9. Fractures. A, Fractures, viewed on bone windows. B, Close-up from A. Fractures are sometimes evident as discontinuities in the bony
cortex. It is important to compare the contralateral side to ensure that a normal suture is not misidentified as a fracture. Air in soft tissues and air–fluid
levels or opacification of sinus air spaces are additional clues to minimally or nondisplaced fractures. (From Broder J, Preston R: An evidence-based approach
to imaging of acute neurological conditions. Emerg Med Pract 9(12):9, 2007.)
abnormal
mastoid normal
air cells mastoid
filled with air cells
fluid are air-
(gray) filled
(black)
A B
Figure 1-10. Mastoiditis resulting in meningitis. This patient presented with confusion and nystagmus, as well as a history of right ear pain.
Noncontrast Computed tomography (CT) showed opacification of the right mastoid air cells, consistent with mastoiditis. Given the patient’s confusion,
meningitis as a complication of mastoiditis was suspected. Cerebrospinal fluid ultimately grew Streptococcus pneumoniae. The mastoid air cells are best
viewed on bone windows (A and B), as soft-tissue windows are subject to bloom artifact from dense bone, which often obscures the small air spaces.
A normal mastoid process is a spongy, air-filled bony lattice. Fluid in the mastoids air cells is abnormal. In the setting of trauma, fluid may indicate blood
filling air cells from a temporal bone fracture. With no history of trauma, infectious mastoiditis should be suspected. Mastoiditis can be recognized on a
noncontrast head CT and does not require special temporal bone views or facial CT series.
When a fracture is identified, look carefully for associ- blood using brain windows. Multiple hemorrhage types
ated brain abnormalities using brain windows. Inspect may coexist. On noncontrast head CT, acute hemorrhage
for any of the types of hemorrhage described later. Look appears hyperdense (brighter or whiter) compared with
for soft-tissue swelling outside the calvarium overlying brain tissue. As time elapses, blood darkens, indicat-
the fracture. ing lower density. This is likely due to a number of fac-
tors, including the absorption of water by hematoma,
B Is for Blood changes in oxidation state, and the dispersion of blood
The second B in our mnemonic is for blood. A brain CT within the subarachnoid space. As discussed later, the
should be carefully inspected for the presence of sub- sensitivity of CT to detect subarachnoid hemorrhage is
arachnoid, epidural, subdural, and intraparenchymal thought to decline as time elapses from the moment of
hemorrhage. Debate exists about the accuracy of CT in Subarachnoid Hemorrhage

dating blood.23 SAH is blood within the subarachnoid space, which
includes the sulci, Sylvian fissure, ventricles, and basilar
Hemorrhage can occur in any of several spaces within cisterns surrounding the brainstem (Figure 1-14).
or around the brain. The shape of blood collections on
CT depends on the anatomic location, as described in Fresh SAH appears white, although the appearance
the sections that follow. varies depending on the ratio of blood to CSF.24 CT is
fracture
no
suture
lines
are
present
fracture
fracture
A B
Figure 1-11. Distinguishing fractures from normal sutures. Fractures, viewed on bone windows. A, Multiple fractures are visible (arrows). The
dotted line connecting to the contralateral side confirms that no sutures are present in these locations. B, Multiple displaced fractures are present (arrows).
The patient’s left maxillary sinus is badly comminuted. The associated air–fluid levels and opacification are additional clues to the presence of fractures.
temporal
bone
fracture
intracranial
air
fracture
air
epidural
hematoma
A B
Figure 1-12. Pneumocephalus, noncontrast CT. A, Brain windows. B, Same image, bone windows. In this patient with a temporal bone fracture
and associated epidural hematoma, air has entered the calvarium and is visible as a small black area on brain windows. On this setting, cerebrospinal
fluid, fat, and and air all appear nearly black, so fine adjustment of the window setting may sometimes be necessary to confirm that this finding is air.
Air may be more visible on bone or lung windows. In this case, the location, adjacent to the fracture, and the rounded shape, typical of air bubbles, are
confirmatory. Finally, the actual density of the tissue can be measured using tools on a digital picture archiving and communication system (PACS). Air
has a density of -1000HU, far lower than any other tissue. (From Broder J, Preston R: An evidence-based approach to imaging of acute neurological conditions.
Emerg Med Pract 9(12):6, 2007.)
fracture
believed to be greater than 95% sensitive for SAH within
the first 12 hours but to decline to 80% or less after
12 hours.25-27 SAH may result from trauma or may occur
spontaneously after rupture of an abnormal vascular
structure such as an aneurysm. When looking for SAH,
inspect the entire subarachnoid space, including the
sulci, ventricles, Sylvian fissure, and cisterns for blood.
air Because subarachnoid blood may diffuse into adjacent
regions, it may defy the guideline that hemorrhage
air and other abnormalities disturb normal brain symme-
try. In other words, large amounts of SAH, including
air hemorrhage into cisterns, may actually result in a sym-
metrical-appearing head CT. Beware of this possibility
when inspecting the brain for abnormalities. Familiar-
ity with the black appearance of normal CSF spaces (see
Figure 1-32) can help to avoid confusion. CSF spaces are
reviewed in detail later with the “C” in our mnemonic.
However, two CSF spaces deserve special mention here
with respect to SAH. The suprasellar cistern lies just
above the sella turcica (home of the pituitary gland) at
the level of the brainstem. This cistern usually is CSF-
filled and has the appearance of a symmetrical black
fracture “star.” When filled with blood, it appears as a symmetri-
cal white star (see Figure 1-14). The quadrigeminal plate
Figure 1-13. Pneumocephalus, noncontrast CT, brain windows. In cistern is usually visible on the same axial CT slice and
this patient with multiple skull fractures, air (arrows) has entered has the appearance of a black “smile” when filled nor-
the calvarium and is visible as black areas on brain windows. On this mally with CSF. When filled with subarachnoid blood, it
setting, cerebrospinal fluid (CSF) and air both appear nearly black, so
fine adjustment of the window setting may sometimes be necessary to becomes a white “smile.”
confirm that this finding is air. Air (-1000 HU) will remain black on all
window settings, while CSF is much denser (0 HU) and will become As time elapses from the moment of hemorrhage, blood
lighter in color with adjustment of the window level. The density of the
tissues can be measured directly using PACS tools when uncertainty will diffuse through the subarachnoid spaces, like a
exists. In this case, the location, immediately adjacent to fractures, and drop of food coloring dropped into a glass of water.
the rounded shape, typical of air bubbles, are confirmatory. Thus a bright white punctate finding on head CT is not
blood in sulci
and Sylvian
fissure
choroid
blood in
pathological choroid blood in posterior horns plexus
quadrigeminal plate
star-sign of plexus of lateral ventricles calcifications
cistern–abnormal
suprasellar calcification
smile sign
cistern blood
caudad cephalad
A B C
Figure 1-14. Subarachnoid hemorrhage (SAH), noncontrast CT, brain windows. Acute SAH appears white on noncontrast computed tomography
(CT) brain windows. A through C, nonconsecutive axial slices, progressing from caudad to cephalad. In this case of diffuse SAH, note the presence of
subarachnoid blood filling the sulci, as well as extending into the cisterns, Sylvian fissures, and even lateral ventricles. In A, blood (white) fills the suprasellar
cistern. This star-shaped structure is normally filled with CSF (black). The quadrigeminal plate cistern is normally a smile-shaped black crescent, filled with
CSF--but in this case is filled with blood. Extremely bright calcifications in the choroid plexus of the posterior horns of the lateral ventricles are common,
normal findings—do not mistake these for hemorrhage. Note their similarity in density to bone of the calvarium.
Box 1-2: CT Findings of Subarachnoid

sulci effaced
Hemorrhage by increased
intracranial
decreased pressure
ll Appearance: White on brain windows size of
ll Location: Localized or diffuse lateral
mm Sulci ventricle
mm Fissures
mm Ventricles
epidural
mm Cisterns hematoma
ll Shape: Assumes shape of surroundings does not cross
ll Pearl: Suspect increased ICP, look for suture line
signs of diffuse edema
ll Pearl: Look for abnormal white “star” sign biconvex
and “smile” sign from SAH in suprasellar epidural
and quadrigeminal plate cisterns hematoma
ventricles
shifted
hetero- across
geneous midline
appearance
likely to be SAH, especially hours after the onset of clini-
cal symptoms. Figure 1-14 shows several examples of
SAH, involving different brain regions. SAH may be
suture
accompanied by other important changes, including depressed
hydrocephalus and cerebral edema, discussed later. Box temporal bone streak
fracture artifact
1-2 summarizes findings of SAH.
Figure 1-15. Epidural hematoma (EDH), noncontrast CT, brain
windows. EDHs are frequently the result of traumatic injury to the
Epidural Hematoma middle meningeal artery at the site of a temporal bone fracture. They
Epidural hematoma (EDH) is a collection of blood characteristically have a biconvex (lens) shape. Because they lie between
lying outside the dura mater, between the dura and the the calvarium and the dura, they are restricted from extending beyond
suture lines, where the dura is tightly adherent to the skull. Sutures are
calvarium. It is almost always a traumatic injury, com- not actually visible on this brain window setting but could be readily
monly resulting from injury to the middle meningeal identified by use of bone windows. Suture locations are pointed out in
artery. Because blood is extravasating from an artery this figure for reference. In this CT, image quality is somewhat degraded
by streak artifact emanating from an object outside of the patient to the
under high pressure, rapid enlargement of the hema- left side (cropped out of the field of image). Despite this, several classic
toma may occur, leading to significant mass effect, features of EDH are visible:
midline shift, and herniation. The common CT appear- • Lenslike or biconvex disc shape
• Temporal location, with associated depressed temporal bone
ance is a biconvex disc or lens, collecting in the poten- fracture
tial space between the calvarium and the dura mater.28 • No crossing of suture lines
This shape occurs because the more superficial aspect • Mass effect with midline shift
• Swirl sign (heterogeneous appearance suggesting active bleeding)
of the EDH conforms to the curve of the calvarium, • Elevated intracranial pressure, with small ventricles and no visible sulci
while the inner aspect expands and presses into the (From Broder J, Preston R: An evidence-based approach to imaging of acute
dura. The dura is usually tethered to the calvarium neurological conditions. Emerg Med Pract 9(12):7, 2007.)
at sutures, so EDHs usually do not cross suture lines
on CT. EDHs may cross the midline, because there are
no midline sutures in the frontal and occipital regions. status or 6-month postoperative status.33 Box 1-3 sum-
The usual location of an EDH is temporal, although marizes findings of EDH.
EDHs occasional occur in other locations. Transfal-
cine herniation may occur with EDHs, so the midline Indications for Surgery in Epidural Hema-
of the brain should be carefully inspected on CT for toma. Published criteria for surgical evacuation of an
midline shift or compression of the lateral ventricle. acute EDH include volume greater than 30 cm3 (regard-
The swirl sign, described as a bright white vortex or less of Glasgow Coma Score, or GCS). Many PACS tool-
“swirl” within the EDH, has long been considered a kits allow automated computation of volumes from
finding of active bleeding and should be interpreted two-dimensional datasets by measuring the thickness
as a sign of continued expansion, although recent of a structure or outlining it. For patients with a GCS
studies have questioned the prognostic significance greater than 8 and no focal deficit, an EDH smaller than
of this finding.29-32 Figures 1-15 and 1-16 show several 30 cm3, less than 15 mm thick, and with less than 5 mm
examples of EDHs, with the classic findings described of midline shift can be managed nonoperatively. Aniso-
earlier. Interestingly, the volume of hematoma has not coria with a GCS below 9 is an indication for surgery,
been shown to correlate with preoperative neurologic regardless of EDH size.34
Box 1-3: CT Findings of Epidural

Hematoma
ll CT Appearance: Variable white to gray on

brain windows
ll Location: peripheral to brain, variable but
usually temporal region
ll Shape: Biconvex disc or lens
ll Pearl: Does not cross suture lines
ll White swirl sign means active bleeding
ll Significance: May cause mass effect and
herniation
mm Look for midline shift
mm Look for effacement of ventricles and sulci
ll Surgical indications34: 15-mm thickness or
5-mm midline shift
SDHs, or acute hematomas in anemic patients, may

biconvex become similar in density (isodense) to the adjacent
epidural no
hematoma visible brain and thus may be difficult to detect.35,36 Clues to
sulci their presence include the obliteration of sulci on the
brain surface and mass effect resulting from the SDH.
Figure 1-16. Epidural hematoma (EDH), noncontrast CT, brain
windows. This EDH (arrowheads) demonstrates several classic features,
Still older SDHs may become similar in density or color
including a lenslike or biconvex disc shape and a temporal location. to the CSF surrounding the brain and thus may be dif-
Although it appears small on this slice, the entire CT may reveal a ficult to recognize. Sometimes SDHs are multicolored or
different story. On this slice, sulci are absent, raising concern for elevated
intracranial pressure.
layered, indicating blood of varying ages. Box 1-4 sum-
marizes findings of SDH.
Subdural Hematoma Indications for Surgery in Subdural Hematoma.

Subdural hematoma (SDH) (Figures 1-17 and 1-18) is Published criteria for surgical evacuation of an acute
a collection of blood between the dura mater and the SDH include thickness greater than 10 mm or midline
brain surface. SDHs usually occur from traumatic injury shift greater than 5 mm, regardless of GCS. Surgery
to bridging dural veins, although a history of trauma is may be indicated with smaller SDHs and lesser degrees
not always found. SDHs may be self-limited in size due of shift in patients with a GCS score less than 9, based
to the lower pressure of venous bleeding, but they can on intracranial pressure (ICP), pupillary findings, and
become enormous, causing significant mass effect, mid- worsening GCS.37
line shift, and herniation. They may also rebleed after
an initial delay, resulting in expansion. Moreover, they Intraparenchymal Hemorrhage
are frequently markers of significant head trauma, and Intraparenchymal hemorrhage (Figures 1-19 and 1-20),
patient outcomes may be compromised by associated dif- or hemorrhage within the substance of the brain mat-
fuse axonal injury (DAI) (described later) or edema. The ter, may occur in trauma or spontaneously, perhaps as a
typical CT appearance of an SDH is a crescent, with the complication of hypertension. The appearance is gener-
convex side facing the calvarium and the concave surface ally bright white acutely. The size may vary from punc-
abutting the brain surface. The shape of SDHs results tate to catastrophically large, with associated mass effect
from their accumulation between the dura and the brain and midline shift. For intraparenchmyal hemorrhage,
surface. Because they lie between the dura and the brain, mass effect such as midline shift or ventricular efface-
they are not restricted by attachment sites between the ment should be assessed. Signs of increased ICP should
dura and the calvarium at sutures. Consequently, SDHs be identified. Particularly for smaller punctate hemor-
may cross suture lines. Moreover, each cerebral hemi- rhages, care must be taken not to mistake hemorrhage
sphere is wrapped in its own dura, so SDHs typically do for normal benign calcifications of the pineal gland, cho-
not cross the midline but instead may continue to follow roid plexus, and meninges, or vice versa. Calcifications
the brain surface into the interhemispheric fissure. can usually be distinguished from hemorrhage as the
former remain bright white on bone windows. In addi-
The color may vary depending on the age of the SDH tion, if the density is measured with PACS tools, calci-
(see Figure 1-18). Fresh subdurals are typically brighter fications have very high density, near +1000HU, while
white (or lighter gray) than the adjacent brain. Older hemorrhage has a density around +40 to +70HU. Pineal
ventricles gland and choroid plexus calcifications also have stereo-

location of
shifted frontal suture typical locations (Figure 1-21) that aid in their recogni-
across
midline
tion. Calcifications are shown in Figure 1-21.
Indications for Surgery in Parenchymal Hemor-

rhage. Published criteria for surgical treatment of trau-
matic intraparenchymal hematoma are complex. They
include size greater than 20 cm3 with midline shift greater
than 5 mm, cisternal compression, or both if the GCS
score is 6 to 8. Lesions greater than 50 cm3 in size should
be managed operatively, regardless of the GCS score.37a
B Is for Brain
midline
shift The third B in our mnemonic is for brain. Brain abnor-
malities include neoplastic masses, localized vasogenic
edema, abscesses, ischemic infarction, global brain
edema, and DAI.
crescent-
shaped Masses
sulci subdural Masses are best delineated on CT with IV contrast or on
effaced hematoma
MRI. However, noncontrast CT can demonstrate a vari-
ety of masses if they are of sufficient size. In some cases,
lambdoid the mass itself may not be seen, but secondary find-
suture
ings such as mass effect, calcification (see Figure 1-21),
Figure 1-17. Subdural hematoma (SDH), noncontrast CT, brain or vasogenic edema (Figure 1-22) (see the discussion
windows. SDHs are most often the result of shearing of dural veins in the next section) may occur. IV contrast is useful in
from blunt trauma. They usually have a crescent shape and may cross detecting masses because they are generally extremely
suture lines, as they lie within the dura and are thus free to extend
along the brain surface, rather than being restricted by the tethering vascular and thus enhance in the presence of contrast
of the dura to the calvarium at sutures. In this figure, the sutures material. On noncontrast CT, masses may be denser
themselves are not visible due to the window setting, but their locations than surrounding brain if they are calcified. Examples
are labeled to illustrate this point. This SDH demonstrates several
classic features: are meningiomas, which are often seen as midline struc-
• Crescent shape tures emanating from the falx cerebri.
• Crossing of suture lines
• Mass effect with midline shift
• Elevated intracranial pressure (ICP), with small ventricles Vasogenic Edema
• No visible sulci (due to effacement from elevated ICP) Malignant primary brain neoplasms or metastatic lesions
(From Broder J, Preston R: An evidence-based approach to imaging of acute often appear hypodense (darker or blacker) compared
neurological conditions. Emerg Med Pract 9(12):5, 2007.)
with normal brain. This appearance is typical of localized
Figure 1-18. Three variations of

subdural hematoma (SDH) from
different patients, noncontrast
CT, brain windows. A, A small
SDH. B, Bilateral SDHs, likely
subacute or chronic, as they are
hypodense relative to brain. C,
Bilateral SDHs with varying density,
likely indicating hemorrhage
at different times. Bright white
hemorrhage is acute, while dark
hemorrhage is subacute or chronic.
Also notice that the window
settings for panels A through C are
not identical. CSF within the lateral
ventricles appears darker in panel A
than in panel C due to differences
in the window setting. Windows
can be varied to accentuate various
tissues. subdural subdural hematomas
hematoma chronic
subdural with blood of varying
hematomas ages
A B C
Box 1-4: CT Findings of Subdural

Hematoma
ll Appearance: Variable white to gray on right lateral

brain windows ventricle
ll Location: peripheral to brain or interhemi-
spheric; variable location around brain
perimeter
ll Shape: Crescent
ll Pearl: May cross suture lines
ll Significance: May cause mass effect and
herniation
mm Look for midline shift
mm Associated with other brain injuries, look
for effacement of ventricles and sulci
intra-
ll Surgical indications37: 10-mm thickness or parenchymal
5-mm midline shift hemorrhage
calcified
mass
vasogenic edema surrounding a lesion. Neoplasms often

secrete vascular endothelial growth factor, resulting in
the development of immature blood vessels that perfuse
the tumor. These immature vessels have leaky endothe-
lial junctions, allowing fluid to extravasate into the inter-
stitium, which causes vasogenic edema. This increased Figure 1-19. Intraparenchymal hemorrhage, noncontrast CT, brain
fluid content reduces the density of brain tissues toward windows. Acute intraparenchymal hemorrhage appears white on CT
that of water (zero on the Hounsfield scale), resulting in brain windows. Hemorrhage in the patient’s left frontal region is creating
mass effect with midline shift. The left lateral ventricle has been completely
a hypodense appearance on CT. In addition, the mass effaced. The single visible ventricle is the right lateral ventricle. A
itself and associated local edema increase the volume of calcified mass in the right occipital region must be differentiated from
brain tissue, resulting in local mass effect, including the acute hemorrhage. Calcifications are extremely bright white on brain
windows—as white and dense as bone. On bone windows, they remain
effacement of ventricles (see Figure 1-22, effacement of white, while hemorrhage does not. The density can also be measured
the posterior horn of the lateral ventricle) and effacement using PACS tools. Bone has a density near +1000HU, while blood
of sulci as adjacent gyri expand in size (see Figure 1-22). has a density closer to +50HU, easily differentiating the two in most
cases. (From Broder J, Preston R: An evidence-based approach to imaging of
acute neurological conditions. Emerg Med Pract 9(12):6, 2007.)
Vasogenic edema must be differentiated from infarction,
which may also cause a hypodense appearance. Vasogenic
edema does not need to conform to a normal vascular ter- clinical scenario. Abscesses, toxoplasmosis, neurocysti-
ritory within the brain, whereas hypodensity associated cercosis (Figure 1-24), and masses all may undergo ring
with ischemic stroke does. Vasogenic edema responds to enhancement, an increase in density around a lesion after
treatment with dexamethasone, whereas steroids are not administration of IV contrast (Box 1-5). This reflects
indicated for other forms of cerebral edema such as trau- increased blood flow in the vicinity of the lesion, as well
matic edema do not. In fact, a multicenter randomized as leaky vascular structures that allow extravasation of
controlled trial of corticosteroid therapy in patients with contrast in the region.
traumatic brain injury showed an increased risk of death
and severe disability from steroid use.38 Ischemic Stroke and Infarction
Ischemic stroke accounts for 85% of strokes.39 It is poten-
As described earlier, midline shift associated with a tially one of the most important indications for head
mass should be carefully assessed during inspection of CT and is an area in which the interpretation of CT by
the brain on brain windows. emergency physicians might play the greatest role by
shortening the time to diagnosis. One obvious reason
Abscesses is the 3-hour or 4.5-hour window for administration of
Abscesses may be visible on noncontrast CT as IV t-PA—an intervention that is still fiercely debated in
hypodense regions (Figure 1-23), occasionally with air the emergency medicine community and that has been
within them. This appearance may be nonspecific, and reviewed elsewhere.40
a differential diagnosis including toxoplasmosis, mass
with vasogenic edema, or central nervous system lym- Understanding CT findings of acute ischemic stroke
phoma should be considered, depending on the patient’s is important—for those who do not believe in
intra-
parenchymal intra-
hemorrhage parenchymal
with mass hemorrhage
A effect B
Figure 1-20. Spectrum of intraparenchymal hemorrhage, noncontrast CT, brain windows. Intraparenchymal hemorrhage may be grossly obvious
or subtle. A, A large hemorrhage with midline shift and mass effect. B, A small amount of hemorrhage (arrow).
normal posterior horn

posterior horn of left lateral
of right lateral ventricle is
ventricle compressed
calcified
meningioma
normal sulci
sulci effaced
vasogenic
choroid edema
plexus mass
Figure 1-22. Masses, noncontrast CT, brain windows. A mass
Figure 1-21. Calcifications, noncontrast CT, brain windows. with surrounding vasogenic edema, which has a hypodense (dark gray)
Calcification of the choroid plexi is a frequent incidental finding that appearance. Neoplasms frequently are associated with vasogenic edema,
may resemble punctate intraparenchymal hemorrhage. Clues are bright named for the putative cause, which is abnormal blood vessels that allow
white density (equal to that of bone), location in the posterior horns extravasation of fluid. This form of edema appears hypodense, like an
of the lateral ventricles, and frequent bilaterality. This patient also has ischemic infarct, but is not restricted to a vascular territory. An abscess
a calcified meningioma. Meningiomas are common benign neoplasms might appear similar. Masses often exert mass effect. Here, the edema
that may become quite large. A well-circumscribed, rounded appearance surrounding the mass is compressing the posterior horn of the left lateral
and calcification are common. (From Broder J, Preston R: An evidence- ventricle. Notice also the effacement of the sulci in the region of the
based approach to imaging of acute neurological conditions. Emerg Med Pract mass. (From Broder J, Preston R: An evidence-based approach to imaging of
9(12):8, 2007.) acute neurological conditions. Emerg Med Pract 9(12):8, 2007.)
surrounding region surrounding

of hypodensity from region of
vasogenic edema hypodensity
(darker gray)
from
vasogenic
parietal lesion edema
lesion
enhances
with IV
A B contrast
Figure 1-23. Brain abscess. This 48-year-old male presented with status epilepticus. CT showed a parietal mass, which at brain biopsy was found
to be an abscess. Cultures grew mixed gram-positive and gram-negative organisms and anaerobes. The patient was subsequently found to be human
immunodeficiency virus positive. A, Noncontrast head CT, brain windows. B, CT with intravenous (IV) contrast moments later, brain windows.
Abscesses and other infectious, inflammatory, or neoplastic lesions typically have surrounding hypodense regions representing vasogenic edema. When
IV contrast is administered (B), the lesion may enhance peripherally, often referred to as ring enhancement.
calcified
edema lesions
ring-
enhancement
edema calcified after IV contrast
lesion administration
cavitary
cysticercosis
lesion with
surrounding
hypodensity of edema B C
Figure 1-24. Neurocysticercosis. This 40-year-old Bolivian male presented with left-hand weakness. A, B, Noncontrast head CT, brain windows. C,
CT with contrast moments later—compare this with image B, a slice through the same level of the brain before contrast administration. Hypodense
lesions are present, with surrounding hypodensity (dark gray) representing edema. Scattered calcifications are also seen, which are a common feature
of old neurocysticercosis lesions. Administration of IV contrast leads to ring enhancement, a feature of many infectious and inflammatory conditions,
including neurocysticercosis, brain abscess, and toxoplasmosis.
TABLE 1-3. Early Ischemic CT Changes Within

Box 1-5: CT Findings of Enhancement 3 Hours of symptom onset, Possibly
Altering Management
ll Increased brightness (Hounsfield units)
after IV contrast administration Type of Change Percentage
ll Seen in Any change 31%
mm Infection
mm Inflammation GWMD loss 27%
mm Neoplasm Hypodensity 9%
mm Vascular lesions CSF space compression 14%
GWMD loss > ⁄3 MCA territory
1
13%
Hypodensity > 1⁄3 MCA territory 2%
administration of t-PA, they provide yet another argu- CSF space compression > ⁄3 MCA1
9%
ment against the treatment, while for those who would territory
use t-PA in select patients, they may allow more rational GWMD, Gray–white matter differentiation.
and safer patient selection. Apart from t-PA administra- From NINDS; Patel SC, Levine SR, Tilley BC, et al: Lack of clinical
tion, rapid diagnosis of ischemic stroke may allow the significance of early ischemic changes on computed tomography in
emergency physician to make better-informed deci- acute stroke. JAMA 286:2830–2838, 2001.
sions about patient management and disposition. If new
stroke therapies such as intra arterial thrombolysis and
clot retrieval become widely accepted and available, Box 1-6: Early CT Findings of Acute
rapid CT interpretation for ischemic stroke may become Ischemia
even more valuable. Interventional radiologic therapies
for ischemic stroke are reviewed in Chapter 16. ll Hyperdense MCA sign
ll Loss of gray–white differentiation
A complex cascade of events occurs to cause the evolv- mm Insular ribbon sign
mm Cortical sulcal effacement
ing appearance of ischemic stroke on head CT. Initially,
ll Ischemic focal hypoattenuation
at the moment of onset of cerebral ischemia, no abnor-
mm MCA territory
malities may be seen on head CT—thus, this is one of the mm Basal ganglia
most difficult diagnoses for the emergency physician, as
a normal CT may correlate with significant pathology. Stud- From Patel SC, Levine SR, Tilley BC, et al: Lack of clinical
ies have shown emergency physicians to be relatively significance of early ischemic changes on computed tomography
poor at recognizing early ischemic changes, which we in acute stroke. JAMA 286:2830-2838, 2001.
review here (Table 1-3 and Box 1-6).
How Early Does the Noncontrast Head CT Indicate this lesion leads to ischemia in the entire MCA territory,
Ischemic Stroke? typically the patient with this finding will have pro-
Analysis of the National Institute of Neurological Disor- found hemiparesis or hemiplegia on the contralateral
ders and Stroke (NINDS) data shows that early ischemic side, as well as other findings such as language impair-
changes are quite common in ischemic stroke, occurring ment, depending on the side of the lesion. In other
in 31% of patients within 3 hours of stroke onset,41 in words, this finding is not associated with mild or subtle
contrast to the widely held belief that ischemic strokes strokes. The presence of a hyperdense MCA sign is an
become visible on CT only after 6 hours.42 Some find- independent predictor of neurologic deterioration.43
ings may occur immediately, such as the hyperdense However,the dense MCA sign is only visible in 30% to
middle cerebral artery (MCA) sign, while other findings 40% of patients with stroke affecting the MCA terri-
may require time to elapse, with the gradual failure of tory.71,72 It may seem surprising that this vascular abnor-
adenosine triphosphate (ATP)–dependent ion pumps mality is visible on noncontrast head CT. As the name
and resulting fluid shifts. implies, the MCA appears hyperdense (bright white)
compared with the normal side. A specific Hounsfield
Hyperdense Middle Cerebral Artery Sign. The unit threshold of greater than 43 units has been recom-
hyperdense MCA sign is a finding of hyperacute stroke, mended to avoid false positives.44
indicating thrombotic occlusion of the proximal MCA.
This may be present on the initial noncontrast head CT Use your knowledge of the location of the patient’s neu-
immediately following symptom onset, since the finding rologic deficits to direct you to the likely side of the lesion,
does not require the failure of ion pumps and fluid shifts which will be on the contralateral side. Then use the
that lead to other ischemic changes on head CT. Because normal symmetry of the brain to help you identify this
normal MCA, hyperdense gray matter

not visible MCA
white matter
Figure 1-25. Hyperdense middle cerebral artery (MCA) sign, Figure 1-26. Normal gray–white matter differentiation, noncontrast
noncontrast CT, brain windows. The hyperdense MCA sign is a CT CT, brain windows. Myelinated regions (white matter) have a greater
finding of thrombosis of the MCA. Importantly, this is a sign seen on fat content than unmyelinated regions (gray matter). As a consequence,
noncontrast CT—the typical initial imaging study obtained in patients and perhaps counterintuitively, white matter has lower density and
with suspected acute stroke. It can be seen in the immediate hyperacute appears darker than gray matter on computed tomography. The dotted
stages of thrombotic–ischemic stroke and may guide therapy, such as line on the patient’s right outlines the border between gray and white
intra arterial tissue plasminogen activator administration. On CT, the matter. Trace this interface yourself on the patient’s left.
hyperdense MCA appears as a white line or point representing the
thrombosed vessel. Care must be taken not to confuse this with the white
appearance of fresh extravascular blood in hemorrhagic stroke. In this
patient, who presented within 30 minutes of onset of right hemiplegia, and white matter is grayer. Figure 1-26 shows the normal
the normal MCA is not visible while the left MCA is thrombosed and
demonstrates the hyperdense MCA sign. (From Broder J, Preston R: An gray–white matter boundary.
evidence-based approach to imaging of acute neurological conditions. Emerg
Med Pract 9(12):12, 2007.) Loss of Gray–White Matter Differentiation. In an
ischemic stroke, as brain tissue consumes ATP and is
unable to replenish it, ATP-dependent ion pumps stop
abnormality. A related finding, the MCA “dot” sign, has working. Ions equilibrate across membranes, and fluid
been validated by angiography and found to be a specific shifts occur. Gray matter gains fluid, lowering its den-
marker of branch occlusion of the MCA. This sign appears sity, and as it does, its density becomes more similar
as a bright white dot in the sylvian fissure on the affected to that of white matter. White matter also gains fluid,
side.45 Figure 1-25 shows the hyperdense MCA sign. increasing its density slightly. Since differences in den-
sity are the reason that these tissues look different on CT,
Gray–White Differentiation. Understanding this fin as their densities converge, their appearances become
ding of stroke requires a brief and simple review of more similar, and it becomes more difficult to discern
neuroanatomy. Gray matter is brain tissue without where gray matter ends and white matter begins. This
myelin—examples include the cerebral cortex, lentiform change is called loss of gray–white matter differentiation,
nucleus, caudate, and thalamus. White matter is com- and it is an early finding of ischemic stroke, occurring
posed of myelinated axons in brain tissue—rendered within 3 hours after onset of ischemia.41 Figures 1-27 and
white on gross pathologic section by the high lipid con- 1-28 show an abnormal gray–white matter boundary.
tent of the myelin sheath. Recall from our earlier discus-
sion of Hounsfield units that lower density on CT means Insular Ribbon Sign (Loss of Insular Ribbon). The
a darker color—low-density fat appears a darker gray insula (or insular cortex) is a thin ribbon of gray mat-
than does higher-density water. Thus, the higher fat con- ter tissue that lies just deep to the lateral brain surface,
tent of white matter makes it appear darker on CT. In separating the temporal lobe from the inferior parietal
other words, on a normal head CT, gray matter is whiter cortex. On CT, it is visible as the tissue layer lining the
Figure 1-27. Early ischemic hypo

density, noncontrast CT, brain
windows. Three examples of
subtle hypoattenuation in early
stroke (dotted circles). Compare the
abnormal side to the normal side
in each image. Large areas of hy-
poattenuation predict an increased
risk of hemorrhage and are relative
contraindications for tissue plas-
minogen activator.
A B C
vasogenic edema both occur. Ion gradients run back

toward equilibrium, and water shifts into gray mat-
ter, making it less dense relative to normal tissue. The
appearance of an infarct becomes progressively more
hypodense over the first several days to weeks of an
ischemic stroke. Again, this finding can occur as an early
change within 3 hours of symptom onset.41 Figures 1-27
through 1-30 show examples of hypodensity. Figure
1-31 shows the progressive hypodensity of an ischemic
stroke over several days.
Hypodensity of Ischemic Stroke Versus Vasogenic

Edema of Masses. Hypodensity in an ischemic stroke
follows a vascular distribution, whereas hypodensity
caused by vasogenic edema around a mass need not
respect vascular territories.
Are Ischemic Changes a Contraindication to t-PA?

Early ischemic stroke findings were not used as exclu-
sion criteria in the NINDS trial, which required only
the absence of hemorrhage on initial head CT.46 How-
ever, multiple studies following NINDS have shown an
increased risk of intracranial hemorrhage, bad neuro-
logic outcomes, and death in patients with early isch-
Figure 1-28. Gray and white matter differentiation, noncontrast CT,
brain windows. When ischemia renders the gray–white interface less emic changes on head CT.47,48 Ischemic changes are
discrete, the computed tomography appearance is called loss of gray– relative contraindications to t-PA administration, and
white differentiation. In this example, the differentiation is normal on their presence may suggest that greater than 3 hours
the patient’s right but is being lost on the patient’s left. The patient has
progressed beyond early hypoattenuation (Figure 1-27) and is developing have elapsed from symptom onset, in which case sys-
the frank hypodensity of ischemic stroke. temic t-PA may be contraindicated. In addition, the Food
and Drug Administration, American Heart Association, and
American Academy of Neurology specifically recommend
Sylvian fissure. This region is subject to early ischemic against administering t-PA if early signs of major infarction
changes in the form of loss of gray–white matter differ- are present, because of increased risk of intracranial hemor-
entiation, often called the insular ribbon sign or loss of rhage.49–51 MCA infarction greater than one third of the
the insular ribbon, as this area becomes less distinct. MCA territory predicts increased bleeding risk if t-PA
is given, and it was poorly detected by radiologists,
Hypodensity in Ischemic Stroke. Ischemic brain neurologists, and emergency physicians in past stud-
looks hypodense, or darker than normal brain in the ies.18,52,53 In addition, the greater the extent of ischemic
same anatomic region. This change occurs for the same changes on CT, the higher the risk of bleeding, as dem-
general reasons as does loss of gray–white differentia- onstrated in the second multinational European Coop-
tion. As neurons deplete stores of ATP, cytotoxic and erative Acute Stroke Study (ECASS-II).48
white matter (myelinated axon tracts)

contain fat, have a low density, and thus
appear darker gray on CT
caudate
lentiform
nucleus
internal
capsule
thalamus
cerebral
cortex
normal
sulci Figure 1-30. Early ischemic changes, noncontrast CT, brain
normal gray- effaced windows. (Same image as Figure 1-29.) Image contrast has been
white matter loss of gray-white sulci increased to accentuate gray–white matter differentiation. On a PACS
differentiation matter differentiation image, you can adjust the window level and contrast. In a normal
brain, you should be able to discern several white and gray matter
Figure 1-29. Early ischemic changes, noncontrast CT, brain
structures. Normal gray–white matter differentiation is subtle. Gray
windows. Early ischemic changes may be visible within 3 hours of onset
matter has lower lipid content than myelinated white matter and
of ischemic stroke. They include sulcal effacement, loss of gray–white
therefore appears brighter on computed tomography (CT). On CT, this
matter differentiation, and the insular ribbon sign. Sulcal effacement
leads counterintuitively to gray matter appearing whiter and white matter
occurs as local edema develops, swelling brain matter and displacing the
appearing grayer. Normal gray matter areas include the cerebral cortex,
cerebrospinal fluid that normally fills sulci as the adjacent gyri become
lentiform nucleus, caudate, and thalamus. White matter tracks, including
edematous. Loss of gray–white matter differentiation occurs as ion pumps
the internal capsule, separate these structures. (From Broder J, Preston R:
fail, leading to equilibration of diffusion gradients and shift of fluid. The
An evidence-based approach to imaging of acute neurological conditions.
normal ability of computed tomography (CT) to differentiate gray from
Emerg Med Pract 9(12):12, 2007.)
white matter relies on differences in their density due to differences
in their fluid and lipid content. White matter contains more fat, is less
dense, and therefore appears darker on CT. Gray matter contains less
lipid, is denser, and therefore appears whiter on CT. Local edema in the differentiation and hypodensity should be identified,
region of a developing infarct renders the region darker on CT because again using the patient’s clinical symptoms to direct
of the presence of increasing amounts of fluid. This masks the normal you to the likely abnormal side of the brain. These early
differentiation between white and gray matter. The insular ribbon sign
is another manifestation of this loss of gray–white matter differentiation. changes may imply either an earlier time of onset than
The insula is a region of gray matter lining the lateral sulcus, in which suggested by the history or a massive stroke in progress.
ischemic strokes of the middle cerebral artery distribution may
demonstrate early abnormalities. In this patient, both sulcal effacement
and loss of gray–white matter differentiation have occurred. The frank Cerebral Edema
hypodensity of ischemic stroke is also becoming visible. Compare these Diffuse cerebral edema is an abnormality of the brain,
to similar regions on the patient’s right side, where normal sulci and but it is manifest through compression of CSF spaces.
normal gray–white matter differentiation are seen. (From Broder J, Preston
R: An evidence-based approach to imaging of acute neurological conditions. It can logically be evaluated under B or C in our mne-
Emerg Med Pract 9(12):12, 2007.) monic. Cerebral edema can result from many pathologic
processes, including trauma, anoxic brain injury, carbon
monoxide poisoning, and systemic fluid and electrolyte
The many noncontrast CT findings of ischemic stroke abnormalities. The appearance is therefore not diag-
may seem too much to hope to remember, and their nostic of the underlying etiology. As the brain swells,
clinical relevance may appear unclear. A few simple several visible changes occur on noncontrast CT. CSF
rules can make sense of this. First, a normal head CT spaces become collapsed as they give way to the increas-
is perhaps the most likely finding if the patient pres- ing volume of solid brain tissue. As a result, the lateral
ents within 3 hours of symptom onset. In this setting, ventricles become slitlike and ultimately become oblit-
the most important job of the emergency physician in erated. In addition, the sulci become effaced as the gyri
interpreting the head CT is to rule out hemorrhage. Sec- swell. The normal rim of CSF surrounding the brain dis-
ond, in the presence of significant unilateral neurologic appears. The cisterns surrounding the brainstem become
abnormalities, the hyperdense MCA sign should be compressed, and risk of herniation rises. Moreover,
sought. Third, early changes such as loss of gray–white as the ICP rises, the cerebral perfusion pressure falls
ischemic ischemic
hypo- hypo-
A density B density
Figure 1-31. Progression of ischemic hypodensity over days, noncontrast CT, brain windows. A left middle cerebral artery distribution stroke,
day 2 (A) and day 4 (B) after symptom onset. Early ischemic changes may be visible within 3 hours of symptom onset. The rate of progression of
computed tomography findings may depend on the degree of ischemia or infarction and thus may vary between patients. Unlike vasogenic edema, this
hypodense region follows a vascular territory and has a wedge-shaped appearance. Also note the local effacement of sulci in the region of the infarct due
to local edema. (From Broder J, Preston R: An evidence-based approach to imaging of acute neurological conditions. Emerg Med Pract 9(12):8, 2007.)
Box 1-7: Relationship Among Intra- Box 1-8: CT Findings of Cerebral

cranial Pressure, Arterial Blood Pres- Edema
sure, and Cerebral Perfusion Pressure
ll Loss of CSF-containing spaces
As ICP rises, blood flow to the brain decreases unless mm Sulci effacement
a compensatory rise in blood pressure occurs: mm Ventricular effacement
mm Cistern effacement
(cerebral perfusion pressure) = ll Loss of gray–white differentiation
(mean arterial pressure) − ICP ll Significance
mm Increased ICP
mm Global brain ischemia from decreased
cerebral perfusion pressure
(in the absence of a compensatory rise in mean arterial
pressure) and global brain ischemia occurs (Box 1-7). Just
as with focal ischemia (stroke), ion pumps fail and loss
of gray–white matter differentiation occurs. Figure 1-35
later in this chapter shows changes of diffuse cerebral Normal Findings That May Simulate Disease
edema. Box 1-8 summarizes findings of cerebral edema. Several common incidental findings may simulate dis-
ease. These include calcifications in the choroid plexus
Diffuse Axonal Injury of the posterior horns of the lateral ventricles (recall that
DAI is the widespread shearing of long axons that occurs the choroid plexus secretes CSF) (see Figure 1-21) and
as the result of deceleration injury. Common clinical sce- calcifications in the pineal gland.55 These should not be
narios include high-speed motor vehicle collisions and confused with hemorrhage, because they have a greater
falls from great height. This injury is not typical of blows density (brighter white appearance) and a stereo typical
to the head or penetrating brain injury. The CT appear- location. The significance of these findings is unknown,
ance is nonspecific: normal in the hyperacute phase, although choroid calcifications have been associated
often followed by cerebral edema over hours to days. with hallucinations in schizophrenia.56
Punctate intraparenchymal hemorrhage may occur as
well. Often, other traumatic brain injury will be evident,
such as SDH or EDH. Why Can Two Patients With the Same CT Findings
Have Markedly Different Neurologic Examinations?
The prognosis is poor, and resolution of CT findings may Remember that CT offers a macroscopic snapshot in
not equate with clinical improvement. MRI is thought to time of complex pathologic changes. It may be that a
be more diagnostic.54 patient with severe neurologic impairment but a relative
normal Figure 1-32. Normal cerebro

quadrigeminal spinal fluid spaces, noncontrast
plate cistern, CT, brain windows. Images are
“smile sign” from the same patient, progressing
sulci from caudad to cephalad. A,
sulci
In a normal brain, the basilar
cistern is patent and filled with
black CSF (sometimes referred
to as the “smile sign”). B, C, The
lateral ventricles are open but not
enlarged. In all panels, sulci are
visible but not excessive. (From
Broder J, Preston R: An evidence-
based approach to imaging of acute
lateral neurological conditions. Emerg Med
ventricles Pract 9(12):5, 2007.)
A B C
Figure 1-33. Cerebral atrophy,

noncontrast CT, brain windows.
enlarged Images are from the same patient,
prominent
lateral
sulci progressing from caudad to
ventricles cephalad. In cerebral atrophy, all
cerebrospinal fluid spaces become
prominent. A, The quadrigeminal
plate cistern is open. A–C, The
lateral ventricles are enlarged and
sulci are prominent, helping to
distinguish this condition from
hydrocephalus, where ventricles
are large but sulci are effaced. (From
normal enlarged based approach to imaging of acute
quadrigeminal lateral
ventricles
neurological conditions. Emerg Med
plate cistern,
“smile sign” Pract 9(12):9, 2007.)
A B C
benign–looking head CT will soon develop changes been found. The CSF spaces offer clues to ICP and may
such as cerebral edema due to neuronal injury that has reveal a neurosurgical emergency. In addition, as dis-
already occurred or is ongoing. In other cases, a patient cussed earlier, SAH may accumulate in CSF spaces,
with a large SDH may appear surprisingly neurologi- including sulci, cisterns, and ventricles. Normal CSF
cally intact, whereas another patient with similar head spaces (Figure 1-32) show symmetrical lateral ventri-
CT findings is severely impaired. One explanation is cles that are neither enlarged nor effaced, patent sulci,
the degree of DAI that may accompany abnormalities and patent basilar cisterns. Deviations from this norm
such as SDH. The patient with minimal deficits may are best appreciated by understanding the normal pat-
have no DAI, whereas the patient with severe deficits tern. In cerebral atrophy (Figure 1-33), all CSF spaces are
may have severe DAI, which is not as evident on CT. enlarged. In obstructive hydrocephalus (Figure 1-34),
Another of many factors that may determine clinical sta- the enlarging ventricles compress other CSF spaces,
tus is the amount of cerebral atrophy present before the causing effacement of the sulci and basilar cisterns. In
injury. Atrophy is loss of brain volume and compensa- diffuse cerebral edema (discussed earlier under “B is for
tory increase in the size of CSF-containing spaces, such Brain”) (Figure 1-35), the swelling brain parenchyma
as ventricles, cisterns, and sulci. When an injury occurs compresses and effaces all CSF spaces, including sulci,
and cerebral edema or a space-occupying lesion such as ventricles, and cisterns. Figure 1-36 compares various
an SDH develops, the presence of atrophy (see Figure combinations of CSF spaces and their diagnostic corre-
1-33) may be protective by allowing room for expansion lates. Table 1-4 summarizes changes in CSF spaces and
of the pathologic lesion without leading to herniation or the accompanying change in ICP.
precipitous rises in ICP.
Because the volume of the calvarium is fixed in patients
C Is for CSF Spaces whose fontanelles and sutures have closed, as the size of
The final letter in our mnemonic, C, reminds us to one component of skull contents (brain, CSF, and blood)
inspect CSF spaces. This is critical, even in cases in which increases, the volume of other components must dimin-
other pathology, such as intracranial hemorrhage, has already ish. Once the compressible skull contents (CSF spaces
quadrigeminal plate enlarged

cistern lateral
effaced ventricles
enlarged
lateral
ventricles enlarged
third
ventricle
sulci
effaced
sulci
effaced
A B C
Figure 1-34. Hydrocephalus, noncontrast CT, brain windows. Images are from the same patient, progressing from caudad to cephalad. In hydro-
cephalus, the quadrigeminal plate cistern is effaced, resulting in loss of the normal smile sign. A–C, Because the total volume of intracranial contents
is fixed in patients whose fontanelles and sutures have closed, as the ventricles enlarge, the sulci become effaced. B, The lateral ventricles and third
ventricle are enlarged. (From Broder J, Preston R: An evidence-based approach to imaging of acute neurological conditions. Emerg Med Pract 9(12):9, 2007.)
Figure 1-35. Cerebral edema, quadrigeminal

noncontrast CT, brain windows. slit-like,
plate effaced
Images are from the same patient, cistern
progressing from caudad to ventricles
effaced
cephalad. A, The quadrigeminal
plate cistern becomes effaced,
resulting in loss of the normal smile
sign. B, C, The lateral ventricles
become compressed and slitlike,
or even completely effaced.
Sulci become effaced. (From
sulci sulci
based approach to imaging of acute
effaced effaced
neurological conditions. Emerg Med
Pract 9(12):9, 2007.)
A B C
and circulating blood volume) have been reduced to enlarged. Because the volume of the calvarium is fixed,
their minimum volumes, ICP rises rapidly if other cra- and solid brain tissue is essentially incompressible, as
nial contents continue to enlarge. the ventricles expand, other CSF spaces become com-
pressed—consequently, the sulci become effaced. In
Sulci contrast, in the patient with atrophy, the ventricles may
Sulci, the CSF spaces between the undulating gyri of the appear dilated but sulci appear similarly enlarged. In
brain surface, should appear black on brain windows. communicating hydrocephalus, the axial CT at the level
Normal sulci are visible but not prominent, and a thin of the lateral ventricles and third ventricle can resemble
ribbon of CSF should outline the entire brain. In cases of the face of a Halloween pumpkin. The enlarged ante-
cerebral edema, the sulci may be completely effaced as rior horns of the lateral ventricles are the rounded eyes,
the brain swells. In hydrocephalus, the volume of brain and the normally slitlike midline third ventricle dilates
remains fixed but ventricles increase in size, leading to to a rounded nose. The posterior horns of the two lateral
compression of the sulci. In cases of cerebral atrophy, ventricles resemble corners of a grin, though they do
loss of brain tissue volume leads to a compensatory not connect in the midline. Depending on the location
increase in the size of sulci (Figures 1-32 through 1-36). of obstruction to CSF flow, the fourth ventricle just pos-
terior to the quadrigeminal plate cistern may become
Ventricles and Hydrocephalus dilated, producing an O-shaped mouth. The quadri-
Hydrocephalus is an important finding for emergency geminal plate cistern itself may be effaced if significant
physicians because of its potential as a neurosurgi- downward herniation is occurring. This appearance has
cal emergency. Untreated, hydrocephalus can result in also been described as a cartwheel, with 5 spokes (the 2
tonsillar herniation, brainstem compression, and respi- anterior and 2 posterior horns of the lateral ventricles,
ratory arrest.57 In general, as hydrocephalus becomes plus the fourth ventricle) radiating from an axle (the
severe, the lateral ventricles become significantly third ventricle). In noncommunicating hydrocephalus,
Figure 1-36. Comparison of cere

brospinal fluid (CSF) spaces.
• Overview of CSF spaces
• Normal brain
• All CSF spaces are present,
normal neither effaced nor enlarged
• Atrophy
• All CSF spaces are enlarged
• Hydrocephalus
• Ventricles expand
• Sulci and cisterns are
compressed
• Edema
• All CSF spaces are
compressed
(From Broder J, Preston R: An
evidence-based approach to imaging of
acute neurological conditions. Emerg
Med Pract 9(12):9, 2007.)
atrophy
hydrocephalus
edema
the obstruction may lie proximal to the fourth ventricle, Computed Tomography Findings of Elevated Intracranial
which will then not appear dilated. Comparison with Pressure
a prior head CT is always valuable in assessing for A variety of CT findings may indicate elevated ICP,
hydrocephalus, because ventricular size alone is a rela- though none are completely predictive. Findings that
tively poor predictor of ICP.58 Normal ventricular size suggest increased ICP are decreased ventricle size,
does not completely exclude the possibility of increased decreased basilar cistern size, effacement of sulci, degree
intracranial pressure. of transfalcine herniation (midline shift), and loss of
gray–white matter differentiation (Box 1-10).59
A variety of CT criteria for acute and chronic hydro-
cephalus have been described (Box 1-9). Figure 1-34
shows CT findings of hydrocephalus. Figure 1-37 shows What CT Findings Are Contraindications to Lumbar
a disconnected ventriculoperitoneal shunt, which can Puncture?
result in hydrocephalus. Before performing lumbar puncture, confirm that find-
ings of midline shift or elevated ICP are not present. The
Atrophy midline should be midline, the sulci should be evident,
In cerebral atrophy, loss of brain volume results in rel- and the cisterns should be open. Ventricle size should
atively symmetrical increase in size of sulci and ven- not be excessive, particularly when compared to sulci.
tricles. The basilar cisterns should also remain patent The evidence basis for assessment of ICP using CT is
(see Figure 1-33). reviewed later in this chapter.
the national average charge for head CT as $996 and for

DETERMINATION OF NEED
MRI as $2283.61 Additional radiologist physician fees
FOR IMAGING may apply. Some authors have concluded that, in the
We have discussed interpretation of head CT in detail, setting of traumatic brain injury, the extreme cost of a
which prepares us for the harder task we face as emer- missed injury justifies the use of CT in all patients, rather
gency physicians: determining which patients actu- than a more selective imaging policy based on clinical
ally require imaging. We begin this assessment with criteria.62 However, with an annual cost of emergency
a review of the costs and radiation risks of imaging, head CT in the United States estimated to exceed $130
because these are important factors in deferring imag- million, others have estimated the savings from selec-
ing when possible. tive use of CT to be high and the risk of missed injury
to be low using validated clinical decision rules such as
Costs of Imaging the Canadian CT Head Rule (CCHR), discussed later in
Costs of CT and MR tests are listed in Table 1-5. These this chapter.63
Medicare reimbursement figures may dramatically
underestimate the cost billed to the patient. An indus- Radiation
try survey of imaging costs in New Jersey found a wide Radiation dose from head CT is approximately 60
variation in consumer costs, ranging from $1000 to mGy.64,65 Attributable mortality risk varies, depending
$4750 for brain MRI or magnetic resonance angiogra- on age of exposure. A single head CT in a neonate would
phy (MRA).60 The American Hospital Directory reports be expected to contribute less than a 1 in 2000 lifetime
risk of fatal cancer, whereas in adults the risk declines
even further, to less than 1 in 10,000.65 However, head
TABLE 1-4. Assessment of Cerebrospinal Fluid
CT may have other risks—one study of patients under-
Spaces and Intracranial Pressure
going external beam radiation therapy for scalp hem-
Condition Sulci Ventricles Cisterns angiomas, with a radiation exposure similar to that
Atrophy Enlarged Enlarged Enlarged from CT, found an association with lower high school
graduation rates.66 This study, although large (more
Hydrocephalus Effaced Enlarged Effaced
than 2000 subjects followed over time), was retrospec-
Cerebral edema Effaced Effaced Effaced tive and therefore can demonstrate only association, not
catheter tip
A catheter tip
B
Figure 1-37. Ventriculoperitoneal shunt failure. A ventriculoperitoneal shunt series is a series of plain x-ray images documenting the course of a shunt
from brain to peritoneum. Rarely, a shunt series may reveal an abnormality despite a normal brain CT. Much more frequently, the shunt series will be
abnormal only if the head CT shows hydrocephalus. A, Chest x-ray from a shunt series. B, Close-up from the same image. In this patient, a discontinuity
is faintly visible between the shunt catheter in the chest and that in the abdomen. The patient’s head CT demonstrated hydrocephalus. (From Broder J,
Preston R: An evidence-based approach to imaging of acute neurological conditions. Emerg Med Pract 9(12):17, 2007.)
Box 1-9: CT Findings of Acute Hydroc Box 1-10: CT Findings of Increased

ephalus59a,59b,59c Intracranial Pressure from Cerebral
Edema59c
ll Dilated lateral ventricles
ll Effaced sulci ll Ventricle size
ll Sylvian and interhemispheric fissures mm Slitlike or none
effaced ll Basilar cistern size
ll Ratio largest width of frontal horns: inter- mm Mildly effaced or effaced
nal diameter from inner table to inner table ll Sulci size
>0.5 mm Effaced or none visible
ll Ratio of largest width of frontal horns to ll Transfalcine herniation (midline shift)
maximum biparietal diameter >30% ll Loss of gray–white matter differentiation
ll Periventricular low density due to transep-
endymal absorption
ll Ballooning of frontal horns of lateral ven-
tricles and third ventricle, also known as TABLE 1-5. Costs of Computed Tomography
“Mickey Mouse” ventricles, and aqueduc- and Magnetic Resonance Tests*
tal obstruction Diagnostic Procedure Cost ($)
ll Cartwheel or jack-o-lantern appearance
CT head or brain without contrast 230.54
CTA head without contrast, followed by con- 382.19
trast, further sections, and postprocessing
causation. In general, the radiation exposure from head
MRA, head with contrast 386.64
CT likely poses a very low level of risk for deleterious
biologic effects, but care should be taken to perform test- MRA, neck with contrast 386.64
ing only when indicated, as radiation effects are cumu- MRI, brain with contrast 386.64
lative and not fully understood. MRA, neck without contrast material(s), fol- 522.54
lowed by contrast material(s) and further
Emergency Department Evaluation sequences
Before imaging can be considered, basic principles of *Based on 2007 Medicare reimbursement national averages.
emergency medicine must be applied, including man- From Reginald D, Williams AR II, Glaudemans J: Pricing Variations in
agement of the patient’s airway and hemodynamic the Consumer Market for Diagnostic Imaging Services. Avalere Health
stabilization as indicated. An unstable patient is not LLC, CareCore National, December 2005; American Hospital Direc-
tory. (Accessed at http://www.ahd.com/sample_outpatient.html.)
appropriate for imaging tests that will take the patient
out of the ED for extended periods, such as MRI. The
history and physical examination can guide imaging test. A complete review of neuroanatomic localization is
decisions. A thorough neurologic examination, includ- beyond the scope of this chapter.
ing assessment of orientation, strength, sensation, deep
tendon reflexes, cerebellar function, and language, may Critical Appraisal of the Literature
help localize the neurologic lesion to assist in choice of A large number of studies have examined indications
imaging modality. Motor and sensory deficits that are for imaging, as well as the test characteristics (including
unilateral may be more suggestive of an anterior fossa sensitivity, specificity, and positive and negative likeli-
brain abnormality, imaged by CT or MRI, whereas a hood ratios) of the available imaging modalities. When
bilateral motor and sensory level may suggest a spinal possible, this chapter focuses on large, multicenter, pro-
lesion. Symptoms of vertigo, ataxia, and dysmetria may spective trials; unfortunately, for many of the clinical
suggest posterior fossa cerebellar abnormalities, best questions addressed, strong evidence is lacking. In the
imaged by MRI. Acute onset of these symptoms could sections that follow, we discuss many studies, address-
suggest posterior circulation stroke, for which imaging ing methodologic strengths and weaknesses. Before we
options would include MRI with MRA and CT angiog- begin, we briefly review some common principles of
raphy (CTA) of the head and neck, described in more evidence-based medicine.
detail later. Symptoms of cranial nerve dysfunction,
including dysarthria, dysphagia, and abnormalities of Principles of Evidence-Based Medicine
extraocular muscles, suggest brainstem pathology bet- for Imaging Studies
ter imaged with MRI than with CT. A motor deficit with Imaging studies for neurologic emergencies share a
ptosis and miosis may suggest carotid artery aneurysm common problem in that the gold standard for diagnosis
or dissection, imaged by CTA, MRA, or carotid ultra- is often another imaging study, with no clear indepen-
sound. Table 1-1 correlates chief complaints, the dif- dent means of settling discrepancies. It is unclear what
ferential diagnosis, and the suggested initial imaging strategy should be used when two imaging studies yield
divergent results. For example, if CT is compared to MR evidence, no level A recommendations could be made
for evaluation of acute intracranial hemorrhage, which for indications for imaging. Level B and C recommenda-
test should serve as the gold standard? Given a nega- tions are listed in Box 1-12. Overall, the sensitivity and
tive CT in the context of a positive MR, is the CT a false specificity of history and physical are limited in identi-
negative or the MR a false positive? Alternative gold fying patients who require emergency imaging.
standards may include clinical follow-up for mortality,
readmission, neurosurgical intervention, or neurologic Imaging for Suspected Stroke
outcome. The most stringent reference standard might Stroke is the leading cause of disability in the United
be autopsy findings, compared with imaging findings. States68 and may be ischemic (85%) or hemorrhagic
When evaluating a study’s relevance to clinical practice, (15%) in nature. When presented with signs and symp-
the strength of the gold standard must be considered. toms suggestive of stroke, the emergency physician must
take steps to differentiate ischemic stroke from intrapa-
Another important concept when interpreting the results renchymal hemorrhage while entertaining the possibility
of a study is point estimate versus 95% confidence interval of stroke mimics (e.g., hypoglycemia or Todd’s paralysis).
(CI). Take the example of a study with a point estimate A number of neuroimaging modalities may aid in this
sensitivity of 99% and a CI of 66% to 100%. The 95% CI task. Some techniques may yield additional informa-
indicates that the true sensitivity of the test in an infi- tion, including the vascular territory affected, the extent
nitely large sample has a 95% chance of lying between of injury, clues to the underlying precipitant cause or
the extreme values of 66% and 100%. Although the like- causes, and identification of tissue that, though ischemic,
lihood of the test having either of these extreme values might still be viable. This information is critical when
is low, it cannot be ruled out on the basis of the data. contemplating the use of thrombolytic or neuroprotec-
Small studies often have broad 95% CIs for their results, tive therapies. Unfortunately, many imaging modalities
whereas large studies usually have narrower CIs. For a are not currently available at all institutions during all
test to be reliable for ruling out a disease process, it must
have both a high sensitivity and a narrow CI. To rule in
pathology, the specificity must be high and the CI nar-
row. The lower boundary of the CI can be considered a
Box 1-12: Guidelines for Imaging in
“worst-case scenario” for the test characteristic. Adult Patients Presenting With Acute
Headache
Another means of reporting a test’s ability to “rule in”
or “rule out” pathology is the likelihood ratio (Box 1-11). ll Level A recommendations: None
The likelihood ratio positive (LR+) is the factor by which ll Level B recommendations
1. Patients presenting to the ED with
the odds of disease increases when the test result is posi-
headache and new abnormal findings in a
tive. The likelihood ratio negative (LR−) is the factor by neurologic examination (e.g., focal deficit,
which the odds of disease decreases when the test result altered mental status, and altered cogni-
is negative. The pretest odds multiplied by the likeli- tive function) should undergo emergent*
hood ratio (positive or negative) yields the posttest odds. noncontrast head CT.
2. Patients presenting with new sudden-
Positive and negative predictive values are not empha- onset severe headache should undergo
sized in this chapter in that they are heavily influenced emergent* head CT.
by disease prevalence and thus must be used cautiously 3. Human immunodeficiency virus–positive
in clinical practice. patients with a new type of headache
should be considered for an emergent*
imaging study.
Which ED Patients With Acute Headache Require ll Level C recommendations
Emergency Imaging? mm Patients who are older than 50 years and
presenting with a new type of headache
The American College of Emergency Physicians (ACEP) but with a normal neurologic examina-
published an update to its clinical policy on evaluation should be considered for an urgent†
tion of acute headache in 2008.67 Based on the available imaging study.
From ACEP; Edlow JA, Panagos PD, Godwin SA et al. Clinical

Box 1-11: Likelihood Ratio policy: Critical issues in the evaluation and management of adult
patients presenting to the emergency department with acute
headache. Ann Emerg Med 52:407-436, 2008.
LR + = sensitivity / (1 − specificity)
*Emergent is defined as CT before ED disposition.
LR − = (1 − sensitivity) / specificity †Urgent is defined as scheduled as part of ED disposition or
oddspost = oddspre × LR performed in ED if follow-up cannot be assured.
hours of the day. Furthermore, many theoretically useful contrast to the assertion in some emergency medicine
options remain untested or inconclusive with regard to texts that ischemic stroke becomes visible on noncon-
their utility in guiding intervention and disposition. trast head CT only after 6 hours,42 the NINDS trial upon
which t-PA therapy is largely based demonstrated that
Computed Tomography for Stroke 31% of patients had early findings of ischemia within 3
CT is the most widely available, immediate imaging hours of symptom onset (see Table 1-3).41 Although early
technique for patients presenting to the ED with signs ischemic changes were not an exclusion criterion for
and symptoms of stroke. Noncontrast head CT is rapid, t-PA in the original NINDS trial,46 subsequent research
taking less than 5 seconds for image acquisition using has shown a heightened risk of hemorrhagic conversion
some 64-slice scanners.22 It is sensitive for detecting of ischemic stroke, poor neurologic outcomes, and death
intracranial hemorrhage,26 and immediate imaging in patients with these changes.47,48 As a consequence,
is more cost effective than either delayed or selective and as mentioned earlier, the Food and Drug Admin-
imaging strategies.69 Limitations do exist, however. Sur- istration, American Heart Association, and American
rounding bone can obscure evidence of ischemic stroke, Academy of Neurology recommend against use of t-PA
an artifact effect known as beam hardening. This prob- in the patients with major early ischemic changes.73 Spe-
lem can be minimized by requesting fine thickness cuts cifically, early ischemic changes occupying an area one
(~1 mm), but the risk of false negatives for stroke detec- third the size of the MCA territory or one third of a cere-
tion still exists—particularly when a vertebral–basilar bral hemisphere, cerebral edema, and midline shift are
distribution is present, because beam hardening is wors- considered relative contraindications to t-PA because of
ened by thick bone surrounding the posterior fossa.70 the increased risk for hemorrhage.74 When discussing
the head CT findings with a radiologist before admin-
Noncontrast CT findings of ischemic stroke were istration of t-PA, it is important to ask specifically about
reviewed earlier in the section on head CT interpreta- the presence and extent of these changes, in addition to
tion. The prognostic importance of these findings and asking about hemorrhage.
implications for t-PA therapy are also discussed there.
Although the sensitivity of noncontrast head CT for Standard contrast-enhanced head CT is rarely used
ischemic stroke increases beyond 24 hours, the sensitiv- for evaluation of stroke since it provides little addi-
ity for ischemia-induced changes in the first six hours is tional information compared with noncontrast head
relatively low at 66%. Thus, a normal CT is consistent CT. With the advent of multidetector CT scanners and
with the diagnosis of acute ischemic stroke in a patient spiral CT technology, however, CT angiography (CTA)
presenting with suggestive signs and symptoms.47 can be performed to obtain images of the extracranial
and intracranial vasculature from the aortic arch to
For the emergency physician, several points about the cranial vertex (Figs. 1-38 through 1-42). Images are
early ischemic changes should be emphasized. First, in acquired by administering a rapid bolus of IV contrast
filling
defect
patent vertebral
partially artery
thrombosed
vertebral artery C1 vertebra
A B vertebral artery
contrast in stenotic region of

true lumen vertebral artery
thrombosed
C false lumen D
Figure 1-38. Computed tomography angiography (CTA). CTA can identify dissections of cervical arteries. A, Axial image. B, Coronal image.
C, Enlarged region from B. D, Three-dimensional reconstruction. In this case, the false lumen has thrombosed, so a filling defect is seen where thrombus
prevents contrast from entering the false lumen. If the false lumen had not thrombosed, an intimal flap would be seen separating the true and false
lumens. Although the three-dimensional reconstruction (D) provides useful anatomic context here, it does not reveal the etiology of the stenosis.
Instead, the thrombosed dissection itself is evident from the cross-sectional images (A–C).
intimal flap
Figure 1-39. Computed tomography angiography (CTA). In this patient with carotid artery dissection, CTA allows three-dimensional reconstruction
of the vessel, with cross sections displayed at intervals around the periphery. An intimal flap is faintly visible on several of these cross sections
(arrows). (From Broder J, Preston R: An evidence-based approach to imaging of acute neurological conditions. Emerg Med Pract 9(12):13, 2007.)
immediately after standard noncontrast head CT. The of exposure to contrast and additional radiation are
raw images can be acquired in as little as 60 seconds, acceptable to the patient.
and three-dimensional computer reconstructions can be
performed in less than 15 minutes. The results might CTA uses enhancement of the cerebral vasculature as a
profoundly alter the course of management, because surrogate for estimating perfusion of the parenchyma.
large artery occlusions correlate with National Insti- Targeted CT perfusion studies (CTPSs) can be performed
tutes of Health Stroke Scale scores75 and may indicate simultaneously using the same bolus of contrast77 and
a need for endovascular intervention (Discussed in have a sensitivity and specificity for detecting ischemia of
Chapter 16). Generally, agreement between CTA and 95% and 100%, respectively.78 By measuring the rise and
catheter angiography—still the gold standard for diag- fall in concentration of injected contrast over time, CTPSs
nosis of vessel stenosis—approaches 95%. For severe are capable of even more direct estimates of cerebral
carotid artery stenosis, sensitivity for CTA approaches perfusion than CTA, including measurements of cere-
100%, whereas the sensitivity for diminished flow in bral blood volume and cerebral blood flow (see Figure
the circle of Willis is 89%.76 Although traditional angi- 1-41). Quantifying these variables may allow clinicians
ography may have subtle, additional benefits related to to identify areas of the brain that, although ischemic, are
characterization of the plaque lesion, the noninvasive potentially still viable—the so-called ischemic penum-
and rapid nature of CTA renders it an attractive option bra. This has implications for the clinician attempting to
to the emergency physician, assuming that the risks weigh the benefits of administering IV or intraarterial
basilar
cisterns
are patent
and
contain
low-
density
CSF
(black),
without
evidence
of sub-
arachnoid
hemor-
rhage,
which
would
appear
white
A B
left MCA aneurysm
5.0 mm
2.9 mm
C
Figure 1-40. Computed tomography angiography (CTA) of intracranial vessels. A, Noncontrast head CT. B, CT angiogram, axial image. C, Three-
dimensional reconstruction rendered from axial CT dataset. CTA can identify berry aneurysms and arterial–venous malformations. This 53-year-old
female awoke at 4 am with a severe headache. Noncontrast head CT was normal, but lumbar puncture showed xanthochromia and more than 20,000
red blood cells per cubic millimeter. CTA confirmed an aneurysm of the left middle cerebral artery (MCA) trifurcation, with one of the MCA branches
arising from the aneurysm. The patient underwent clipping of the aneurysm rather than angiographic coiling because of this configuration.
use of thrombolytics in patients presenting after an arbi-

trary time interval (e.g., 3 hours), thrombolytic therapy
could be initiated or excluded based on actual visual-
ization or absence of a penumbral area likely to benefit
from such intervention.
However, routine use of CTPSs in the hospital setting

(much less in the ED) is not without challenges. First, the
usual difficulties of imaging the posterior fossa with CT
techniques persist.70,80 Second, only limited volumes of
brain can be imaged at one time with each bolus of con-
trast, so ischemia located outside the scanning level of
interest can be missed,81 although this is partially allevi-
ated by the use of multislice scanners or repeated contrast
Figure 1-41. Computed tomography (CT) perfusion scan. A CT boluses. Third, despite the theoretical appeal, only small
perfusion scan provides a map of blood flow in ischemic stroke, potentially studies in limited populations exist that confirm the abil-
identifying ischemic areas that could be salvaged by reperfusion. Here, ity of CTPSs to detect infarct,82 to predict infarct location83
blood flow in similar regions of the two cerebral hemispheres is compared.
This is an imaging technique that continues to be developed in the research and size,84 and to predict final outcomes.82,84 We await
setting but is not a routine part of most clinical practice. Erroneously large study confirmation of these findings before rou-
programmed CT perfusion scans have been the cause of iatrogenic radiation tinely recommending their use in the ED setting. Because
injury to patients undergoing stroke evaluations. (From Broder J, Preston R:
An evidence-based approach to imaging of acute neurological conditions. Emerg CT perfusion studies require that the same regions of the
Med Pract 9(12):13, 2007.) brain be repeatedly scanned over a period of a few min-
utes to acquire data about perfusion over time, focal areas
thrombolytic drugs against the risk of intracranial hem- of the brain receive higher levels of radiation exposure.
orrhage. Routine use of CTPSs could potentially allow Widely publicized accounts exist of patients receiving
a more precise prediction of outcome79 and could even dangerous radiation exposures when CT scanners were
herald a paradigm shift in one of the indications for misprogrammed, prompting lawsuits and an investiga-
thrombolytic administration: rather than excluding the tion by the US Food and Drug Administration.84a
A C
B D
Figure 1-42. Multimodality assessment of stroke. This patient presented with dizziness, nausea, and vomiting. Noncontrast computed tomography
of the brain (A) was interpreted as normal, but symptoms were concerning for posterior circulation stroke. CT angiography (B) suggested right vertebral
artery dissection (arrow). Magnetic resonance imaging (C) revealed posterior inferior cerebellar artery–territory ischemic stroke (arrows) and magnetic
resonance angiography (MRA) (D) confirmed vertebral dissection (arrow). E, A three-dimensional view from the MRA. (From Broder J, Preston R: An
evidence-based approach to imaging of acute neurological conditions. Emerg Med Pract 9(12):15, 2007.)
Magnetic Resonance Imaging for Acute Stroke information was designated as “highly likely” to affect
Standard MRI for stroke includes scout images, T1- and management strategy in 38%.89
T2-weighted images and MRA. Increasingly available
new-generation scanners incorporate additional high- MRI is superior to CT at both detecting acute ischemic
sensitivity methods such as diffusion-weighted imag- change85,87 and visualizing the posterior fossa.70,80
ing (DWI), gradient echo pulse sequencing (GEPS), and However, MRI has failed to supplant CT as the imag-
perfusion-weighted imaging (PWI). ing modality of choice for stroke in the ED due to cost,
availability of the requisite personnel, time, and a long-
Obtaining DWI has been possible since 1985.85 In brief, standing belief that MRI is not reliable for detecting
the technique involves detecting and processing a signal intracerebral hemorrhage. At least the last two factors
in response to the movement of water molecules caused are being surmounted. New scanners are faster, with
by two pulses of radiofrequency. Ischemic changes can acquisition times in the range of 3 to 5 minutes, com-
be detected in as little as 3 to 30 minutes after insult, and pared to 15 to 20 minutes previously.90 With respect to
in a small study of 22 patients who presented within hemorrhage, DWI has proved sufficient to exclude intra-
6 hours of symptom onset, DWI was found to be 100% cerebral hemorrhage,91 and in studies comparing GEPS
sensitive and specific.86 In a subsequent study, DWI was with CT, GEPS was at least as useful for detecting acute
found to have a far superior sensitivity compared to intracranial hemorrhage and actually better at elucidat-
CT (91% vs. 61%).87 When MRA is done simultaneous ing chronic hemorrhagic changes,92-94 with sensitivity
with DWI as part of a fast protocol to detect vascular approaching 100% when interpreted by trained person-
stenosis, their combined use within 24 hours of hospi- nel.92 The issues of cost and personnel are more com-
talization substantially improved the early diagnostic plex, however. MRI hardware costs and costs associated
accuracy of ischemic stroke subtypes.88 DWI may also with imaging are roughly double those of CT.95 Whether
be useful when the clinician encounters a patient with these costs will fall in the future or can be justified in the
remote-onset neurologic defects. In patients presenting form of better outcomes, shorter hospital stays, or other
with a median delay of 17 days after symptom onset, measurable end points is unknown. Personnel issues are
clinicians gained additional clinical information one related not only to sheer manpower but also to qualita-
third of the time (including clarification of the vascu- tive training demands. MRI requires specially trained
lar territory affected) by performing DWI in addition technicians spending more time per study compared to
to conventional T2-weighted images. Of this third, the CT, and expert-level radiologists with extensive training
in MRI interpretation must be employed, because inter- to be illuminated), and duplex (a combination of the
pretation is still not reproducible (though advocates of two). Carotid duplex ultrasound has traditionally been
DWI point out that there is virtually no intra-observer or deployed electively (i.e., nonemergently) to investigate
interobserver variability with this modality).87 whether the origin of an acute ischemic event in a given
patient could be due to carotid artery stenosis. Studies
Similar to the ability of CTPSs to identify the ischemic show conflicting results, with some showing poorer per-
penumbra, the combination of DWI and PWI makes it formance of ultrasound (65% sensitivity, 95% specificity)
possible to make inferences about ischemia before per- compared with MRA (82%-100% sensitivity, 95%-100%
manent injury (infarction) has occurred. DWI provides specificity) and the gold standard (by definition 100%
a map of brain tissue that is ischemic and at high risk sensitive and specific) of digital subtraction angiogra-
for infarction. PWI provides a map of brain tissue that is phy.106,107 Transcranial ultrasound can be used to visu-
threatened by decreased blood flow but not yet demon- alize the vessels in and near the circle of Willis. Here, it
strating cellular injury marked by changes in diffusion. is possible to identify stenosis with reasonable success,
Typically, the hypoperfused region (PWI defect) is larger though less well for the vertebral–basilar system. In the
than the area of diffusion abnormality (DWI defect) internal carotid artery distribution, the sensitivity and
early in stroke. The DWI–PWI mismatch is thought to specificity for stenosis are 85% and 95%, respectively.
represent the ischemic penumbra, a region that is poten- Sensitivity and specificity are only 75% and 85%, respec-
tially salvageable with aggressive reperfusion (either by tively, in the vertebral–basilar system. Additional ben-
thrombolytic therapy or by catheter-based mechanical efits of transcranial ultrasound reportedly include the
interventions).96 ability to identify collateral pathways, visualize in real
time harmful emboli and in the postthrombolysis state,
PWI is performed with standard MRI and MRA using judge the success of therapy.108–110 In addition, ultra-
gadolinium and requires a total imaging time of less than sound is being used therapeutically in trials to augment
15 minutes. PWI can be performed in cases of contrain- the thrombolytic effect of medications.111
dication to gadolinium (a rare event, as gadolinium has
been found safe in most instances, though recent fatal Studies have focused on the use of ultrasound to select
nephrogenic systemic fibrosis has been noted in patients patients for thrombolytic drug therapy or endovascular
with advanced renal disease; see Chapter 1597), by mag- treatment. Ultrasound is inexpensive relative to other
netically labeling the blood as the blood enters the brain, imaging modalities, noninvasive, and does not expose
a technique known as continuous arterial spin label- the patient to ionizing radiation, but the validity of the
ing.98 Patients with large perfusion defects96,99 detected results is highly operator-dependent. In addition, it is
by PWI or occluded arteries detected by MRA100 are at impossible to differentiate reliably between complete
heightened risk for enlarging regions of frank infarc- and high-grade stenosis, and contralateral stenosis
tion, leading some to suggest that these findings should can result in falsely reassuring flow velocities ipsilat-
prompt early revascularization, either with thrombolytic erally, resulting in a false-negative interpretation for
agents pharmacologically or with mechanical devices. stenosis.112–115
The volume of abnormalities on DWI and PWI during
acute stroke correlates with acute National Institutes of Conventional Catheter Angiography for Stroke
Health Stroke Scales and with chronic neurologic scores, Still considered the gold standard for diagnosing arte-
and lesion size may predict early neurologic deteriora- rial stenosis, conventional catheter angiography has
tion.101–102 Still another benefit of advanced techniques been substantially improved with the advent of digital
like DWI and PWI is, as is the case for CTPSs, the poten- subtraction techniques permitting visualization of even
tial to identify areas of ischemia that have not yet pro- small, cortical branches of intracranial arteries. Endo-
gressed to infarction, potentially permitting extension of vascular techniques permit administration of intra-
the traditional 3-hour window for thrombolytic admin- arterial thrombolysis, and some users have the ability to
istration.103,104 However, PWI and DWI have yet to prove perform clot retrieval and angioplasty with or without
practical and reliable in defining the ischemic penumbra stenting.116 However, despite its utility when noninva-
and infarct core,105 and head-to-head trials comparing sive diagnostic techniques are equivocal or conflicting,
MRI diffusion–perfusion studies to CTPSs are limited. angiography is still used only sparingly in the acute
stroke setting because of its invasive nature and the
Figure 1-42 shows images from a single patient in vari- approximate 1% risk of iatrogenic stroke associated with
ous modalities, including CT, CTA, MRI, and MRA. the procedure.117
Ultrasonography for Ischemic Stroke Stroke Neuroimaging Summary

Ultrasound techniques include Doppler (used to assess The goals of neuroimaging in the ED patient present-
flow rate and presence of stenosis), brightness mode ing with signs and symptoms consistent with stroke
(permitting anatomic and structural details of the tissue include the exclusion of intraparenchymal hemorrhage,
space-occupying lesions, and other stroke “mimics.” Ide- the ED, CT is not necessary as MR can evaluate for hem-
ally, the imaging technique would also highlight areas of orrhage, infarct, and other brain pathology.
ischemia, identify underlying causes of ischemia (i.e., ves-
sel occlusion), and delineate those areas that are ischemic
but salvageable. Currently, no single imaging modality Is MRI Useful in Apparent TIA?
can accomplish all of these goals quickly and at low cost, MRI may be a useful alternative imaging modality when
and no combination of studies is widely available in all available for patients with apparent TIA. The age of isch-
centers. emic lesions can be judged with DWI based on measure-
ment of the apparent diffusion coefficient, which varies
When stroke is the suspected diagnosis, the patient must with time from onset of ischemia. Ischemic lesions of
be imaged as quickly as possible with the most readily varying ages on DWI MRI predict a substantially higher
available neuroimaging modality to rule out hemorrhage risk of new lesions on 30-day follow-up MRI (relative
and other stroke mimics. Typically, this is with noncon- risk = 3.6; 95% CI = 1.9-6.8), so MRI may be useful for
trast CT. In the absence of contraindication to the use of risk stratification.122 Ischemic lesions of varying ages
contrast, simultaneous CTA can evaluate for large vessel suggest an embolic source of TIA and future stroke.
occlusion. Provided that it does not delay other indicated A negative diffusion weighted MRI within 24 hours of
therapy, CTPS may be useful for prognosis and to guide apparent TIA predicts a low risk of disabling ischemic
therapeutic decisions, including the use of thrombolytic stroke within 90 days in patients with moderate or high
drug therapy. In specialized centers with the required risk ABCD(2) scores (described later).122a
expertise and resources, MR stroke protocols including
MRI, MRA, DWI, and PWI may be feasible from the ED
without the need for prior CT imaging. What Percentage of Patients with TIA or Stroke
Have a Cardioembolic Source? Is Echocardiogram
Clinical Questions in Stroke Imaging Indicated?
Studies show that 51% to 61% of patients with TIA or
What Is the Risk of Progression to Stroke in Acute stroke in whom a clear cause had not yet been identified
TIA? What Imaging Studies Are Indicated? after initial workup had a cardiac abnormality poten-
Large studies in multiple settings have demonstrated tially warranting anticoagulation identified on echo-
that patients diagnosed with TIA in the ED progress to cardiography. Transesophageal echocardiography was
stroke at a high rate of 10% in 3 months, with half of those superior to transthoracic echocardiography, identifying
strokes occurring within 48 hours.118,120 In addition, an abnormality in 40% of those with a normal transtho-
some evidence suggests that patients with TIA may be racic echocardiography. An absolute indication for anti-
at higher risk for subsequent adverse clinical outcomes coagulation was identified by echocardiography in 20%
than patients with minor ischemic strokes—possibly of patients, and 80% of those were identified only on
because of a higher rate of large vessel atherosclerosis transesophageal echocardiography.123,124 Another study
as the cause of TIA. In TIA, large artery atherosclerosis demonstrated that coexisting cardiac and carotid artery
(including carotid artery disease) may account for up to lesions are quite common in TIA and stroke patients,
34% of cases. The 3-month risk for stroke, myocardial occurring in 11%, so imaging of both the heart and the
infarction, and vascular death is higher for TIA patients cervical vessels is likely indicated.125
than for minor stroke patients (15% vs. 3%; hazard ratio
= 4.6; 95% CI = 2.3-9.3 in multivariate analysis).121 The
result has been a significant impetus to admit patients Does a Clinical Prediction Rule Exist to Identify
with TIA or to perform urgent diagnostic testing in the High-Risk TIA Patients Who Require Further
outpatient setting. A variety of neuroimaging strategies Imaging?
have been proposed to identify patients at high risk. The ABCD score has been described to identify patients
at low risk of progression to TIA. The rule incorporates
age, blood pressure, unilateral weakness, speech distur-
Is Noncontrast CT Valuable in Apparent TIA? bance, and duration to generate a score, which has been
In patients presenting with apparent TIA, head CT correlated with 7-day risk for stroke. A dichotomized
might be expected to be normal, and the value of CT version of the ABCD score, with patients scoring five
in the context of resolved neurologic symptoms could or more points being at high risk, has been validated
be questioned. Yet 4% of patients clinically diagnosed for the ED. Unfortunately, the small numbers in these
with TIA in the ED have evidence of new infarct on non- studies result in wide CIs for the sensitivity of the score,
contrast head CT. Moreover, those patients have a four- with lower 95% CIs as low as 60%. Some studies have
fold increased 90-day risk of subsequent stroke.118 Thus, shown the ABCD score to have only limited value, as
head CT may be warranted in TIA as a risk stratification patients with relatively low risk scores (predicted to cor-
tool. If MRI with or without MRA will be performed in relate with low risk for stroke) had adverse outcomes
Box 1-13: ABCD(2) Score for Suspected Subarachnoid Hemorrhage

redicting Progression from Transient
P Traditional practice in the evaluation for suspected SAH
Ischemic Attacks to Stroke* has been lumbar puncture (LP) following negative CT, a
practice still advocated by major textbooks of emergency
ll Age ≥60 years (1 point) medicine.133 The reported sensitivity of CT (third gen-
ll Systolic blood pressure ≥140 mm Hg eration or higher) for SAH is in the range of 90% in the
and/or diastolic blood pressure ≥90 mm Hg first 24 hours, declining afterward.134 A recent study of
(1 point) fifth-generation CT found no SAH in patients undergo-
ll Unilateral weakness (2 points) ing LP after negative CT.27 However, this retrospective
ll Speech disturbance without weakness
review examined only 177 ED patients undergoing both
(1 point)
CT and LP. Records and follow-up were not reviewed
ll Duration of symptoms
mm ≥60 minutes (2 points) for patients who underwent CT but not LP, so it is pos-
mm 10–59 minutes (1 point) sible that cases of SAH with negative head CT occurred
mm <10 minutes (0 points) but were not detected. In addition, the interval between
ll Diabetes (1 point) headache onset and CT or LP was not recorded, so this
study provides no information on any time dependency
From Asimos AW, Johnson AM, Rosamond WD, et al. A multi- of CT sensitivity for SAH. Given an incidence of SAH of
center evaluation of the ABCD2 score’s accuracy for predicting only around 3.4% compared with previously reported
early ischemic stroke in admitted patients with transient isch-
numbers in the range of 12% of ED acute, severe, non-
emic attack. Ann Emerg Med 55:201-210 e5, 2010; Bray JE,
Coughlan K, Bladin C. Can the ABCD score be dichotomised to traumatic headache patients, the true sensitivity of
identify high-risk patients with transient ischaemic attack in the fifth-generation CT may be as low as 61%.26,27,135 A pro-
emergency department? Emerg Med J 24:92-95, 2007; Cucchi- spective study of patients presenting with the worst
ara BL, Messe SR, Taylor RA, et al. Is the ABCD score useful headache of their life reported a sensitivity of 97.5%,
for risk stratification of patients with acute transient ischemic
but again the small number of patients, 107, yielded a
attack? Stroke 37:1710-1714, 2006; Johnston SC, Rothwell
PM, Nguyen-Huynh MN, et al. Validation and refinement of lower CI—as low as 91%.26 Another retrospective study
scores to predict very early stroke risk after transient ischaemic found the sensitivity of noncontrast head CT for SAH
attack. Lancet 369:283-292, 2007; Rothwell PM, Giles MF, to be 93% (95% CI = 88%-97%).136 Sensitivity of CT is
Flossmann E, et al. A simple score (ABCD) to identify individu- thought to decline over time due to clearance of blood
als at high early risk of stroke after transient ischaemic attack.
and is reported to be as low as 50% at 7 days.137 How
Lancet 366:29-36, 2005; Tsivgoulis G, Spengos K, Manta P,
et al. Validation of the ABCD score in identifying individuals good is the combination of CT with LP for ruling out
at high early risk of stroke after a transient ischemic attack: the diagnosis of SAH? A prospective cohort study of 599
A hospital-based case series study. Stroke 37:2892-2897, 2006. patients undergoing both LP and head CT for nontrau-
*In validation studies, ABCD(2) poorly predicts short-term matic acute headache found the combination of head
stroke risk.
CT and LP to be 98% sensitive (95% CI = 91%-100%) for
the diagnosis of SAH or aneurysm. The authors calcu-
lated a negative likelihood ratio of 0.024, indicating an
extremely low posttest probability of SAH when both
in external validation.126–129 A recent extended version CT and LP are negative.138 Emergency medicine physi-
of the ABCD score including diabetes as a risk factor, cians are only moderately accurate at predicting SAH
known as ABCD(2), attempts to predict 2-day risk for based on clinical history,139 so a conservative approach
stroke but requires further validation (Box 1-13).130 including LP after negative head CT is probably still
A multicenter study demonstrated a low 7-day risk for warranted based on CT sensitivity. Even the authors of
progression from TIA to disabling stroke in patients studies of CT sensitivity are reluctant to state that LP is
with an ABCD(2) score of no more than three. How- not needed after negative CT.140 Future advances in CT
ever, risk for more minor stroke was poorly predicted, may eliminate this need, although some have argued on
limiting applicability in the ED.131 Another prospective theoretical grounds that CT sensitivity will never reach
study suggested a 5-fold higher 90-day risk of stroke in 100% for SAH.141
patients with an ABCD(2) score >2.131a
CT Angiography for Aneurysmal Subarachnoid
Hemorrhage
What Is the Yield of CTA, Ultrasound, or MRA When noncontrast head CT is negative in a patient
in TIA? suspected of SAH, is additional imaging warranted to
As stated earlier in the discussion of stroke, CTA has assess for aneurysmal disease? Is there a role for CTA
been shown to be 100% sensitive for high-grade carotid (see Figure 1-40)? The precise role is as yet undefined. A
stenosis. MRA and carotid ultrasonography are alterna- small study found aneurysms in 5.1% (6/116) of patients
tive studies with similar sensitivity.132 with a negative CT and positive LP and in 2.5% (3/116)
following a normal CT and LP. Given the incidence of and specificity, greater than 95%.156 Imaging of the head
berry aneurysms in the general population, believed and neck should be ordered when these diagnoses are
to be approximately 1% to 5%,142,144 it is possible that suspected to ensure that the lesion is within the imaged
the aneurysms detected in these patients were inciden- field (see Figures 1-38 and 1-39).
tal, not the acute cause of the patients’ symptoms. Wide
use of CTA might result in detection of large numbers Imaging in Acute Hydrocephalus and Shunt
of asymptomatic aneurysms, resulting in unneeded pro- Failure
cedures, including formal angiography and endovas- Noncontrast head CT is the initial study of choice for
cular coiling of aneurysms, with associated morbidity diagnosis of acute obstructive hydrocephalus. As the
and mortality. Perhaps the best role of CTA would be in ventricles enlarge, due to obstruction of the normal
patients in whom LP is not feasible—for example, those outflow of CSF, other CSF spaces become progressively
with coagulopathy. CTA might also be useful in patients effaced, because the total volume of all skull contents
with a particularly high pretest probability of disease is fixed (Figures 1-34, 1-36, and 1-37). Large ventricles
but negative noncontrast CT and LP.145 Finally, CTA with small or completely effaced sulci and cisterns are
could be used to evaluate for aneurysm in patients with consistent with hydrocephalus. In contrast, in cerebral
negative noncontrast CT and equivocal LP results, such atrophy all CSF spaces are enlarged, whereas in cerebral
as a declining but nonzero number of red blood cells. edema all spaces are effaced (see Figure 1-36). Occasion-
ally, a mixed picture can occur, with both obstructing
hydrocephalus and diffuse cerebral edema. A number
How Is Cerebral Dural Sinus Thrombosis best of radiographic criteria for hydrocephalus have been
Imaged? described (see Box 1-9).
Dural sinus thrombosis is a rare but important cause
of headache. An observational study of 1676 consecu- How Common Is Ventricular Shunt Obstruction
tive Pakistani patients undergoing a CT or MRI study
Among Children Presenting With Suspected
found dural sinus thrombosis in 3.3%, although the
incidence in U.S. ED patients may be quite different.146 Obstruction Undergoing CT? How Sensitive
Risk factors include hypercoagulable states. Untreated, and Specific Are Computed Tomography and Shunt
the condition leads to rising ICP due to failure of drain- Series for Shunt Obstruction?
age of blood from dural sinuses into the internal jugu- Among children presenting with suspected shunt mal-
lar venous system. This in turn can lead to intracranial function, up to 25% may require surgical intervention.
hemorrhage. Both CT venography and MR venography The sensitivity of noncontrast head CT is reported to
can be used to detect this condition. Only small stud- be 83%, with a specificity of 76%. Traditionally, a shunt
ies dating from the 1990s have directly compared the series (a series of plain x-rays without contrast following
two techniques, and without a strong gold standard, it the course of the shunt from head to destination, usu-
is impossible to state the sensitivity of the techniques ally peritoneum) is also performed. Shunt series have
accurately.147,148 The technique for CT venography is the low sensitivity (20%) but high specificity (98%). Studies
same as for CTA with one exception: in CTV, CT image do not suggest high utility of this test in patients with
acquisition is timed to coincide with the arrival of con- normal noncontrast head CT, but occasionally the shunt
trast in venous structures.149,152 series may be positive in these patients (see Figure 1-37).
In one series, 3 of 233 patients undergoing imaging had
Imaging of Vascular Dissections normal head CT but abnormal shunt series and docu-
Vascular dissections of the carotid and vertebral arteries mented shunt obstruction.58,157
are a relatively rare cause of acute headache and neu-
rologic symptoms, occurring in an estimated 2.5 to 3 When head CT and standard shunt series are normal
per 100,000 and 1 to 1.5 100,000 annually in the United but shunt malfunction continues to be suspected, MRI
States.153,154 They account for only about 2% of all isch- can be used to assess flow within the shunt. The tech-
emic strokes but as much as 20% of strokes in young nique is felt to be highly sensitive but not perfectly spe-
adults.155 Noncontrast head CT is expected to be negative cific. Some patients may have intermittent flow through
in the setting of these lesions unless ischemic stroke has a functioning shunt, and if MRI is performed during a
resulted. In addition, dissection of the vertebral arteries period of normal physiologic low flow, the MRI may
would be expected to result in ischemia in the region falsely suggest the shunt to be occluded. However, dem-
of the basilar artery (which forms from the confluence onstration of flow by MRI confirms shunt patency.158,159
of the vertebral arteries), and the territories supplied by
the basilar artery lie within the posterior fossa, an area Central Nervous System Infections
poorly seen on noncontrast CT. Multiple imaging tech- When central nervous system infection is suspected,
niques are available for this diagnosis. CTA, MRA, and imaging may be indicated for diagnosis. Imaging
conventional angiography all have excellent sensitivity serves several roles in this setting: it evaluates for other
diagnoses, including hemorrhage or masses; it identifies carefully considered in patients with abnormal mental
focal infectious processes, such as abscess or toxoplas- status or neurologic examinations, even if CT appears
mosis; and it identifies possible contraindications to LP, normal. Conversely, significant midline shift or findings
such as the presence of elevated ICP. suggesting herniation on CT may rarely be present in
patients with normal neurologic examinations. Follow-
Assessment of Intracranial Pressure Before Lumbar ing head trauma, 1.9% of patients with CT-diagnosed
Puncture: Does CT predict elevated ICP before LP? frank herniation and 4.4% of patients with significant
Some studies have shown size of ventricles on CT to brain shift but no herniation had no neurologic deficit
be only weakly predictive of ICP.58 Oliver et al. have in National Emergency X-radiography Utilization Study
argued that the evidence that imaging accurately pre- (NEXUS) II.162
dicts elevated ICP is scant, that some degree of ICP
elevation in meningitis is probably ubiquitous, and that Imaging in Patients With Seizures
examination findings suggesting high ICP, such as stu- Imaging of patients with new-onset seizure who have
por, coma, or focal neurologic deficits, are more reliable returned to a normal neurologic baseline is a level B
than CT in identifying patients at risk of herniation.160 recommendation in the 2004 ACEP clinical policy on
Nonetheless, CT is frequently performed before LP for seizures (Box 1-14).163 Level B recommendations gener-
suspected meningitis, with the goal of ruling out alter- ally reflect evidence with moderate certainty based on
native diagnoses and identifying contraindications to class II studies (e.g., nonrandomized trials, retrospec-
LP. A prospective study in 2001 demonstrated an asso- tive or observational studies, and case-control studies)
ciation among age, recent seizure, abnormal neurologic directly addressing a clinical question or reflect broad
examination findings, and immunocompromise and CT consensus among experts based on class III studies (e.g.,
abnormalities before LP, although the majority of the case reports and case series). What is the basis of the
CT abnormalities were felt to be unlikely to contraindi- ACEP recommendation? Studies on patients with new-
cate LP (Table 1-6).161 Because elevations of ICP may be onset seizure show a wide range of abnormal head CT
present that are not detected on head CT, LP should be results, from 3% to as high as 41%, often unsuspected on
the basis of history or examination.164,166 It is not clear
whether there is a causal relationship between some of
TABLE 1-6. Predictors of Head CT Abnormalities the CT abnormalities found in these studies and acute
Potentially Contraindicating Lumbar seizure or whether any beneficial change in manage-
Puncture in Patients With Suspected ment resulted from CT imaging. A systematic review
Meningitis
Risk Ratio
Baseline Patient Characteristic (95% CI) Box 1-14: Indications for Imaging
Age ≥60 years 4.3 (2.9-6.4) in Adult Patients With New-Onset
Immunocompromised state* 1.8 (1.1-2.8) Seizure
History of central nervous system disease† 4.8 (3.3-6.9)
Urgent neuroimaging is recommended for all adult
Seizure within 1 week before presentation 3.2 (2.1-5.0)
patients, even with return to baseline neurologic
Neurologic Findings status. Imaging should be performed in the emergency
Abnormal level of consciousness 3.3 (2.2-4.4) department for patients with:
ll Age >40 years
Inability to answer two questions correctly 3.8 (2.5-5.8) ll AIDS or immune compromise
Inability to follow two commands correctly 3.9 (2.6-5.9) ll Anticoagulation
Gaze palsy 3.2 (1.9-5.4) ll Cancer history
ll Fever
Abnormal visual fields 4.0 (2.7-5.9) ll Focal onset before generalization
Facial palsy 4.9 (3.8-6.3) ll New focal deficits
ll Persistent altered mental status
Arm drift 4.0 (2.7-5.8)
ll Persistent headache
Leg drift 4.4 (3.0-6.5) ll Suspected acute intracranial process
Abnormal language‡ 4.3 (2.9-6.5) ll Trauma
From Hasbun R, Abrahams J, Jekel J, Quagliarello VJ: Computed
tomography of the head before lumbar puncture in adults with sus- From ACEP Clinical Policies Committee, Clinical Policies
pected meningitis. N Engl J Med 345:1727-1733, 2001. Subcommittee on Seizures: Clinical policy: Critical issues in the
*Human immunodeficiency virus or AIDS, immunosuppressive therapy, evaluation and management of adult patients presenting to the
or transplant. emergency department with seizures. Ann Emerg Med 43:605-
†Mass lesion, stroke, or focal infection. 625, 2004.
‡Aphasia, dysarthria, or extinction.
found that among all adult ED patients with new-

onset seizure, 17.7% had an abnormal head CT—26.6% Box 1-15: Simple Versus Complex
of those undergoing head CT. Among patients with Febrile Seizure
acquired immunodeficiency syndrome (AIDS) and new-
onset seizure, the risk appears higher still (20%-30%), ll Simple febrile seizure (imaging not gener-
with frequent CT diagnoses including cerebral toxo- ally indicated)
plasmosis, progressive multifocal leukoencephalopa- mm age 3 months to 5 years
thy, and central nervous system lymphoma.167 Whether mm generalized
mm duration <15 minutes
seizure is new onset or recurrent, emergent CT scanning
mm do not recur within 24 hours
is recommended for patients with any of the risk fac-
ll Complex febrile seizure (imaging may be
tors in Box 1-14.168 Noncontrast CT is the initial imaging indicated, though recent studies suggest
method, as the majority of life-threatening lesions (hem- low risk of emergent intervention)
orrhage, edema, mass effect, and hydrocephalus) would mm age <3 months or >5 years
be expected to be found with this modality. Enhanced mm focal, with or without secondary
CT or MRI may be indicated if unenhanced CT is nor- generalization
mal and suspicion remains of a structural lesion. mm duration >15 minutes
mm recur within 24 hours
Febrile and Typical Recurrent Seizures
Neither emergent nor urgent imaging is recommended From Greenberg MK, Barsan WG, Starkman S: Neuroimaging
in the emergency patient presenting with seizure. Neurology
for patients with a typical simple febrile seizure (Box 47:26-32, 1996.
1-15) or recurrent seizures with typical features com-
pared with the patient’s prior seizure history.168 Do
complex febrile seizures require emergency imaging?
A recent study found that although 16% of children of patients presenting to a community ED underwent
presenting with new-onset complex febrile seizure had head CT, with only 1 patient (0.7%) having the etiology
abnormal CT or MR findings, none required emergent identified by imaging (posterior circulation infarct).172
intervention (0%; 95% CI = 0%–4%).169 A second retrospective study found that 283 of 649
(44%) patients admitted with syncope at two commu-
Do partial seizures suggest a focal structural brain nity teaching hospitals from 1994 to 1998 underwent
abnormality and thus mandate imaging? A study from head CT, with 5 (2%) revealing an apparent causal diag-
South Africa in a region with a high incidence of tuber- nosis: 10 patients underwent MRI without a diagnosis of
culosis and neurocysticercosis suggests that the answer a cause of syncope. Utilization of head CT fell from 61%
is no. This prospective cohort study of 118 children with to 33% of syncope patients from 1994 to 1998, but diag-
new-onset partial seizure found abnormal CT scans in nostic yield remained extremely low, under 1% for both
only 8 (7%), all of whom were suspected prospectively groups.173 A prospective observational study found a
of having the CT diagnosis. The investigators con- 39% rate of head CT in ED syncope patients at Harvard
cluded that routine scanning would require 11 scans University’s Beth Israel Deaconess Medical Center, with
and 5 courses of antihelmintic therapy to prevent one only 5% having head CT abnormalities. That research
case of childhood seizure disorder, versus no scans and group proposes that a decision rule for head CT in syn-
11 courses of drug therapy if all seizure patients were cope might reduce utilization by 25% to 50%, although
empirically treated.170 Of course, this study is from a such a rule has not been prospectively validated (Box
population quite different from the U.S. population, 1-16).174 Head CT may be warranted when the history
and its results must be viewed in this context. and exam do not fully exclude other diagnoses, such as
seizure or stroke, or when trauma results from a syn-
Head and Brain Imaging in Syncope cope episode.175 A 2007 ACEP Clinical Policy found that
Head CT is commonly obtained as part of the evaluation there is no evidence for routine screening of syncope
of a patient with syncope, despite little evidence sup- patients with advanced imaging including CT. That
porting its use. Syncope is a brief, nontraumatic loss of policy gives a Level C recommendation for cranial CT
consciousness with loss of postural tone. Syncope is due only when indicated by specific findings in the history
to global cerebral hypoperfusion, but the brain is gener- or physical examination.175a
ally the “victim” of syncope, not the cause. Syncope is
rarely due to stroke, as loss of consciousness requires Posterior Reversible Encephalopathy Syndrome
loss of blood flow to both cerebral hemispheres or to the (PRES)
medullary reticular activating system in the brainstem. Posterior Reversible Encephalopathy Syndrome (PRES)
Studies of the diagnostic yield of head CT performed is a state of vasogenic brain edema seen in severe hyper-
for syncope show little pathology clearly related to the tension, eclampsia, lupus, and cyclosporine toxicity,
syncope episode.171 In one retrospective study, 34% among many other conditions. The mechanism remains
Box 1-16: Proposed Decision Rule Box 1-17: National Emergency

for Computed Tomography Head X-radiography Utilization Study II:
in Syncope* Intracranial Injuries Considered
Significant
ll CT head indicated only for patients with
one or more of the following ll Mass effect or sulcal effacement
mm Signs or symptoms of neurologic disease, ll Signs of herniation
including headache ll Basal cistern compression or midline shift
mm Trauma above the clavicles ll Substantial EDHs or SDHs (>1.0 cm in
mm Coumadin use width or causing mass effect)
mm Age >60 years ll Substantial cerebral contusion (>1.0 cm in
diameter or >one site)
From Grossman SA, Fischer C, Bar JL, et al: The yield of head ll Extensive SAH
CT in syncope: A pilot study. Intern Emerg Med 2:46-49, 2007. ll Hemorrhage in the posterior fossa
*Not yet validated. ll Intraventricular hemorrhage
ll Bilateral hemorrhage of any type
ll Depressed or diastatic skull fracture
ll Pneumocephalus
in debate, and studies of imaging findings are limited ll Diffuse cerebral edema
by the lack of a strong diagnostic reference standard. ll DAI
CT and MR can demonstrate focal regions of symmet-
ric hemispheric edema. The most commonly involved From Mower WR, Hoffman JR, Herbert M, et al: Developing a
decision instrument to guide computed tomographic imaging of
regions are the parietal and occipital lobes, followed by
blunt head injury patients. J Trauma 59:954-959, 2005.
the frontal lobes, inferior temporal-occipital junction,
and cerebellum. Vascular watershed areas of the brain
appear most affected. The basal ganglia, brain stem, and
deep white matter (external/internal capsule) can also blunt head trauma, regardless of GCS score or neuro-
be involved. Complications such as hydrocephalus and logic examination. Of these, 6.7% to 8.7% had clinically
hemorrhage can occur. On MRI, restricted diffusion rep- significant traumatic blunt head injury (prospectively
resenting infarction is seen in 11% to 26% of cases. The defined) (Box 1-17).177,178 For purposes of NEXUS II, a
clinical importance of imaging for the specific diagnosis clinically significant traumatic brain injury was defined
and management of PRES is not well established, though based on prior research as an injury that may require
imaging to rule out hemorrhage, mass or mass effect, neurosurgical intervention (such as craniotomy, inva-
hydrocephalus, and ischemic stroke is indicated.175b sive ICP monitoring, or mechanical ventilation) or an
injury with potential for rapid deterioration or long-
Imaging of Traumatic Brain Injury term neurologic dysfunction.179 Based on NEXUS II, an
For evaluation of traumatic brain injury, noncontrast CT average emergency physician evaluating patients with
remains the primary imaging modality. Brain injuries a range of GCS scores and neurologic examination find-
can include traumatic SAH, SDH, EDH, intraparenchy- ings might expect to find a potentially important head
mal hemorrhage, DAI, and traumatic cerebral edema. CT abnormality in between 1 in 20 and 1 in 10 patients
Concurrent injuries may include bony injuries such as in whom CT head was ordered after blunt trauma.177–178
skull fracture. More rarely, head and neck trauma may Stiell and collaborators in Canada showed a 6% inci-
result in vascular dissection of the intracranial or extra- dence of head injuries in ED patients undergoing CT
cranial arteries (carotid or vertebral), with potential cat- following blunt head injury, although they also demon-
astrophic neurologic outcomes such as ischemic stroke. strated great heterogeneity in the ordering practices of
ED physicians, from 6.5% to 80% of patients with head
Epidemiology of Traumatic Brain Injury trauma undergoing CT, depending on treating physi-
The incidence of an acute intracranial injury seen on CT cian.180 As described later, a variety of decision rules
following a “mild” traumatic brain injury (GCS score = have been investigated to reduce unnecessary imaging
13–15) is approximately 6% to 9%, but not all detected in patients with a very low risk for intracranial injury.
injuries result in a clinically meaningful change in man-
agement.63,176 Of patients in the derivation phase of the Skull Films for Blunt Head Trauma
Canadian CT Head Rule (CCHR), 8% had potentially The 2008 ACEP clinical policy on altered mental status
important CT head findings, yet only 1% underwent a and mild blunt head trauma found that “plain films of
neurosurgical intervention.63 NEXUS II enrolled 13,728 the skull have essentially no utility” in the emergency
patients at 21 medical centers in the United States, evaluation of patients with altered mental status.181
including all ED patients undergoing head CT after Surprisingly, they are still widely used in international
TABLE 1-7. Outcomes and Test Characteristics of Three Clinical Decision Rules for Computed Tomography
Use Following Minor Blunt Head Injury in Adults*
GCS or Neurologic Sensitivity and
Examination Time to Imaging Outcome of Interest Specificity (95% CI)
NEXUS-II Any GCS or any examina- Within 24 hours of injury Clinically important 98.3% (97.2-99.0)
tion with or without LOC traumatic injuries 13.7% (13.1-14.3)
New Orleans With LOC, GCS = 15, Within 24 hours of injury Any intracranial injury 100% (95-100)
Criteria normal neurologic 25% (22-28)
examination†
CCHR GCS = 13-15 with LOC, For patients with GCS <15, Injuries requiring 100% (92-100)
amnesia, or confusion 2-hour postinjury observation neurosurgical 68.7% (67-70)
required to observe intervention
for improvement in GCS
Clinically important brain injury 98.4% (96-99)
49.6% (48-51)
Data from Haydel MJ, Preston CA, Mills TJ, et al: Indications for computed tomography in patients with minor head injury. N Engl J Med 343:
100-105, 2000; Mower WR, Hoffman JR, Herbert M, et al: Developing a decision instrument to guide computed tomographic imaging of blunt head
injury patients. J Trauma 59:954-959, 2005; Stiell IG, Wells GA, Vandemheen K, et al: The Canadian CT Head Rule for patients with minor head
injury. Lancet 357:1391-1396, 2001.
*Patients with evidence of more severe head injury, defined by GCS lower than 13 or focal neurologic deficits, require head CT.
†Normal neurologic exam is normal strength, sensation, and cranial nerves.
LOC, loss of consciousness.
emergency practice. In the United Kingdom, skull radio-

graphs are obtained in about 20% of blunt head injury Box 1-18: New Orleans Criteria:
patients, a decrease from nearly 50% in the past.182 Some Positive” Computed Tomography
“
investigators have argued that this diagnostic test may Findings
play a limited role when CT is not available.183–184
ll Any acute traumatic intracranial lesion,
Clinical Decision Rules for Blunt Head Trauma including:
Clinical decision rules have been derived by several mm SDH
mm EDH
groups in the United States, Canada, the United King-
mm Parenchymal hematoma
dom, and Scandinavia with moderate success. These
mm SAH
rules differ in their definitions of clinically significant mm Cerebral contusion
injuries, the neurologic inclusion criteria (GCS score, mm Depressed skull fracture
loss of consciousness, and neurologic examination),
and the time from injury to imaging. Table 1-7 com- From Haydel MJ, Preston CA, Mills TJ, et al: Indications for
pares three rules, while Boxes 1-19 through 1-23 and computed tomography in patients with minor head injury. N Engl
Table 1-8 list the specific inclusion and exclusion criteria J Med 343:100-105, 2000.
and specified outcomes of interest. In general, the New
Orleans Criteria (NOC) (Box 1-18) seek to identify any
acute intracranial injury, while NEXUS II and the CCHR sensitivity would be unacceptable. Their rule achieved
seek to identify injuries most likely to require neuro- this goal of sensitivity (100%; 95% CI = 95%-100%) but
surgical intervention or to result in serious neurologic has a low specificity (25%; 95% CI = 22%-28%).176 This
deficits. It is a matter of debate as to which definition is rule (see Box 1-20) has been challenged for its lack of
most appropriate. For example, is it important to know specificity, which might lead to increased utilization of
of the existence of a traumatic brain injury that does not head CT for patients with no other indication for imag-
require neurosurgical intervention to follow cognitive ing beyond headache, vomiting, or minor head and
function long-term? Moreover, some clinical interven- neck trauma such as facial abrasions.
tions may not have truly been “needed” but rather may
have been driven by the judgment of individual physi- Canadian Computed Tomography Head Rule. The
cians once they became aware of imaging findings. CCHR was 100% sensitive (95% CI 92%-100%) and 68.7%
specific (95% CI 67%-70%) for patients with blunt trauma
New Orleans Criteria. The New Orleans Criteria and a GCS score of 13-15 in its original study; subse-
investigators sought a rule with 100% sensitivity, citing quent studies have found slightly lower values.63,185–189
prior surveys of emergency physicians indicating that This rule (see Box 1-21) has been criticized for its relative
a clinical decision rule with anything less than perfect complexity, as well as for end points (see Box 1-19) that
Box 1-19: Canadian Head CT Rule: Box 1-20: New Orleans Criteria
Outcomes
Head C is required for blunt trauma patients with
Primary Outcomes: Neurosurgical loss of consciousness, GCS score of 15, a normal
Intervention neurologic exam,* and any of the following:
ll Death within 7 days secondary to head injury ll Headache
ll Any of the following procedures within ll Vomiting
7 days: ll Age >60 years
mm Craniotomy ll Drug or alcohol intoxication
mm Elevation of skull fracture ll Deficits in short-term memory
mm ICP monitoring ll Physical evidence of trauma above the
mm Intubation for head injury (demonstrated on clavicles
CT) ll Seizure
Secondary Outcomes
From Haydel MJ, Preston CA, Mills TJ, et al: Indications for
ll Clinically important brain injury on CT
computed tomography in patients with minor head injury.
mm Any acute brain finding revealed on CT N Engl J Med 343:100-105, 2000.
and that would normally require admission *Normal cranial nerves and normal strength and sensation in the
to hospital and neurosurgical follow-up arms and legs, as determined by a physician on the patient’s
ll Not clinically important arrival at the ED.
mm Patient was neurologically intact and had
one of the following:
nn Solitary contusion <5 mm in
diameter
nn Localized subarachnoid blood Box 1-21: Canadian CT Head Rule*
<1 mm thick
nn Smear SDH <4 mm thick For patients with a GCS score of 13-15 after
nn Closed depressed skull fracture witnessed traumatic loss of consciousness, definite
not through the inner table post-traumatic amnesia, or witnessed post-traumatic
disorientation, CT is only required for patients with any
From Stiell IG, Wells GA, Vandemheen K, et al: The Canadian one of the following findings:
CT Head Rule for patients with minor head injury. Lancet High Risk for Neurosurgical Intervention
357:1391-1396, 2001. ll GCS score <15 at 2 hours after injury
ll Suspected open or depressed skull
fracture
ll Any sign of basal skull fracture†
might be considered unacceptable in some medical prac-
ll Two or more episodes of vomiting
tice settings, including the United States, where fears of ll Age ≥65 years
litigation might make any CT abnormality undesirable Medium Risk for Brain Injury Detection
to be missed. Nonetheless, in multiple validation stud- by Computed Tomography
ies and subanalyses, the rule appears to perform well ll Amnesia before impact >30 minutes
in identifying patients who present with normal mental ll Dangerous mechanism‡
status but require emergent neurosurgical intervention,
including in U.S. populations.190 Remarkably, despite *Exclusion criteria: no history of trauma, GCS <13, age <16
numerous studies in multiple settings, emergency phy- years, warfarin use or coagulopathy, obvious open skull fracture.
†Including hemotympanum, raccoon eyes, cerebrospinal fluid,
sician awareness of the rule remains low. A recent study
otorrhea or rhinorrhea, and Battle’s sign.
found that only 35% of U.S. emergency physicians were ‡A pedestrian struck by a motor vehicle, an occupant ejected
aware of the rule (see Table 1-8).191 from a motor vehicle, or a fall from ≥3-foot elevation or five
stairs.
National Emergency X-radiography Utilization From Stiell IG, Wells GA, Vandemheen K, et al: The Canadian
Study II. The NEXUS II investigators identified a deci- CT Head Rule for patients with minor head injury. Lancet
357:1391-1396, 2001.
sion rule with high sensitivity (98.3%; 95% CI = 97.2%–
99.0%) but low specificity (13.7%; 95% CI = 13.1%–14.3%)
for significant intracranial injury.177 This rule (see
Box 1-22) has limited clinical utility because it would II rule instead of current clinical practice would iden-
mandate brain CT in the majority of patients follow- tify important injuries that would otherwise have been
ing blunt trauma and thus might actually increase CT missed. Another difficulty with this proposed rule is the
use when compared with current physician practice. Its potential variability in application of terms such as scalp
clinical outcome benefit is uncertain—no study to date hematoma or abnormal behavior by different observers. In
has demonstrated whether application of the NEXUS addition, coagulopathy, found to be a high-risk factor,
TABLE 1-8. Canadian Computed Tomography Head

Box 1-22: National Emergency Rule: International Survey Responses
X-radiography Utilization Study II:
Australasia Canada United United
Variables Associated With Significant
Kingdom States
Head Injury
Aware of 83% 87% 63% 35%
CCHR
ll Evidence of significant skull fracture
ll Scalp hematoma Use of 55% 83% 44% 29%
ll Neurologic deficit CCHR
ll Altered level of alertness Consider 67% 84% 33% 77%
ll Abnormal behavior future use
ll Coagulopathy (includes aspirin use) From Eagles D, Stiell I, Clement C, et al. An international survey of
ll Persistent vomiting emergency physicians’ knowledge, use, and attitudes towards the
ll Age ≥ 65years Canadian CT Head Rule. Acad Emerg Med 2007;14:S86.
From Mower WR, Hoffman JR, Herbert M, et al: Developing

a decision instrum ent to guide computed tomographic imag-
ing of blunt head injury patients. J Trauma 59:954-959, 2005. would increase CT use and associated costs.193,194 In
German populations, the rules appear to reduce utili-
zation.195 A prospective Dutch study validated the high
sensitivity of both rules but found that the New Orleans
Criteria would decrease use by only 3% and the CCHR
Box 1-23: Prediction Rule for Head by 37.3%.186 These reductions in use are smaller than
Injury in Children Under 2 Years of the estimates from the original New Orleans Criteria
Age publication (20%)176 and CCHR (50% to 70%).63 These
findings reflect the existing practices in other countries;
CT is not indicated for patients meeting these criteria. when baseline CT use for blunt head injury is low, the
ll No scalp hematoma except frontal NEXUS, Canadian, and New Orleans rules may result in
ll No loss of consciousness or loss of con- smaller decreases in use or could actually increase use.
sciousness for less than 5 seconds
ll Normal mental status
ll Nonsevere injury mechanism Is One Rule “Best”?
ll No palpable skull fracture A retrospective study of 7955 patients with traumatic
ll Acting normally according to the parents head injury published in 2009 compared the three rules
we’ve discussed in detail, plus the guidelines of the Neu-
From Kuppermann N, Holmes JF, Dayan PS, et al: Identifica-
rotraumatology Committee of the World Federation of
tion of children at very low risk of clinically-important brain
injuries after head trauma: A prospective cohort study. Lancet Neurosurgical Societies, the National Institute of Clini-
374:1160-1170, 2009. Severe mechanism of injury: motor cal Excellence (UK), and the Scandinavian Neurotrauma
vehicle crash with patient ejection, death of another passenger, Committee. No statistical difference was found when
or rollover; pedestrian or bicyclist without helmet struck by a comparing the sensitivity of the rules for detection of
motorized vehicle; falls of more than 1.5 m [5 feet]; or head
intracranial hematoma requiring surgical treatment.
struck by a high-impact object.
Current ACEP (2008) guidelines continue to endorse the
New Orleans Criteria with a level A recommendation
and the CCHR with a level B recommendation.196
included aspirin use, potentially increasing the need for
head imaging among patients who do not otherwise
appear at risk for brain injury. One important finding Can We “Mix and Match” the Rules?
of NEXUS II is a lack of association among several clini- It may be tempting to adopt a combination of features
cal variables that have traditionally been considered as of the preceding decision rules to manufacture a supe-
potential markers of significant clinical injury, includ- rior decision rule. Unfortunately, there is little evidence
ing loss of consciousness, seizure, severe headache, and to support this practice. The NEXUS investigators and
vomiting. the New Orleans investigators tested a variety of other
combinations of clinical criteria besides the final sug-
External Validity of Decision Rules for Blunt Head gested rules. 176,177 Eliminating criteria not surprisingly
Trauma. The CCHR and the New Orleans Criteria improved specificity but impaired sensitivity, and add-
have been prospectively tested in other populations ing additional criteria improved sensitivity but impaired
with less success; in Australia, the rules fail to exclude specificity. A rule with large numbers of criteria fails the
injury while reducing use.192 In Britain, the CCHR original purpose of developing a clinical decision rule;
it detects all injuries by mandating CT for all blunt head predictors, any of which would mandate CT: abnormal
trauma patients. mental status, clinical signs of skull fracture (including
retroauricular ecchymosis, CSF otorrhea or rhinorrhea,
Some Commonalities Among the Rules and palpable fracture), history of any vomiting, scalp
An important take-home point for all of the decision hematoma (in children ≤2 years of age), or headache.
rules described earlier is the notable absence of loss of The sensitivity of the rule was 99% (95% CI = 94%-
consciousness alone as an indication for head CT. In the 100%) for traumatic brain injury and 100% (95% CI =
CCHR and New Orleans Criteria studies loss of con- 97%-100%) for detection of injury requiring acute inter-
sciousness was a required inclusion criterion, but in vention. This rule would have eliminated 24% of the
none of the studies did isolated loss of consciousness CT scans performed at the study institution, but the CIs
identify patients at risk for significant injury. NEXUS II for the result remain wider than would be accepted by
included patients with or without loss of consciousness some clinicians and parents.
in the study population, and did not find that loss of
consciousness required head CT in the absence of the In the largest study of pediatric blunt head injury to
other rule critera. Prior studies had shown little asso- date, Kuppermann et al.204 conducted a prospective
ciation between loss of consciousness and significant study of 42,412 children ages 18 years and younger to
traumatic brain injury.197 Neither NEXUS II nor the derive and validate a clinical decision rule identifying
CCHR found posttraumatic headache to be an indica- children with extremely low risk for clinically impor-
tion for head CT. The New Orleans Criteria did identify tant traumatic head injuries after blunt trauma.204 Of
headache as a high-risk criterion, though a decision rule these, 14,969 (35.3%) underwent head CT, with clini-
omitting headache would have had 97% sensitivity in cally important head injuries occurring in 376 (0.9%),
the derivation phase.176 Because the New Orleans inves- and neurosurgery occurring in 60 (0.1%). The investiga-
tigators desired 100% sensitivity, they did not further tors identified two rules: one for children younger than
validate such a rule. Single posttraumatic seizure also 2 years and a second for children 2 years and older. For
does not appear to be a high-risk feature by the CCHR children under 2 years of age, the rule had a negative
or NEXUS II, though again, it is considered high risk by predictive value of 100% (95% CI = 99.7%-100%) and
NOC. sensitivity of 100% (95% CI = 86.3%–100%). In children
2 years and older, the negative predictive value was
Table 1-9 summarizes the rules for ease of comparison 99.95% (95% CI = 99.81%-99.99%) and sensitivity was
and application. 96.8% (95% CI = 89.0%-99.6%). The rules would have
eliminated the need for imaging in 24.1% of children
Decision Rules for Children younger than 2 years and 20.1% of older children, with-
Clinical decision rules for children following blunt head out missing any children requiring neurosurgery. Thus
injury remain controversial.198 Predictors of intracranial children meeting all low-risk criteria do not require a
injury such as scalp hematoma lack specificity, result- head CT. Conversely, children who do not meet all of
ing in large numbers of head CT performed to detect the low-risk criteria do not necessarily require an immediate
an injury. For example, only about 1% of patients under head CT; the rules had positive predictive values of less
the age of 2 years with scalp hematoma have traumatic than 2.5% for clinical important traumatic brain injury.
brain injury, although large hematomas, nonfrontal This underscores a major limitation of clinical decision
location, and accompanying skull fracture increase the rules: high sensitivity inevitably comes at the price of
risk.198 In general, acceptably high sensitivity in these poor specificity and potential for overuse. While the
studies comes at the price of extremely low specific- Kuppermann rules identify children with very low risk
ity, resulting in little reduction in the utilization of CT of important brain injury, they are not intended to iden-
in pediatric populations. In some settings, application tify children at high risk. The authors consequently rec-
of these rules could actually increase utilization. These ommend CT or clinical observation in those patients not
rules may also be very sensitive to the care with which meeting all low-risk criteria.
they are applied; the NEXUS II rule, for example, falls
in sensitivity from approximately 99% to only 90% In the youngest patients, preverbal children 0 to 3 years
when the word headache is replaced with severe headache, of age, clinical assessment is particularly difficult. A
so emergency physicians must scrupulously apply the very low threshold for CT must be applied after any
rules if they expect the rules to function as described in head trauma, especially considering the possibility of
the original studies.199-203 nonaccidental trauma.
Palchak et al.200 prospectively derived a clinical decision Decision Rules for Elderly
rule for pediatric patients using a cohort of 2043 patients The elderly have a high rate of injury with few clini-
at a single center. They excluded patients with trivial cal predictors.205 NEXUS II found that the rate of clini-
trauma such as falls from standing. They identified five cally significant injury in those 65 years of age or older
TABLE 1-9. Criteria for Three Clinical Decision Rules for Computed Tomography Use Following Blunt Head
Injury in Adults*
CCHR† New Orleans Criteria NEXUS II
CT is only required for patients with GCS Head CT is required for blunt Head CT is required for
of 13-15 after witnessed traumatic loss of trauma patients with loss of blunt trauma patients with
consciousness, amnesia, or confusion and consciousness, GCS 15, a any of the following:
any one of the following findings: normal neurologic exam,‡ and
High Risk for Neurosurgical Intervention any of the following:
Age • < 16 years (exclusion criteria in original • <3 years (excluded from • No lower limit
study) original study) • ≥65 years
• ≥ 65 years • >60 years
Vomiting • Two or more episodes • Even one episode • Recurrent (no specific
number of episodes)
Traumatic • Suspected open or depressed skull • Physical evidence of trauma • Evidence of significant
findings on fracture above the clavicles skull fracture
examination • Any sign of basal skull fracture§ • Headache • Scalp hematoma
Neurologic • GCS <15 at 2 hours after injury • Deficits in short-term • Altered level of alertness
examination memory • Abnormal behavior
• Drug or alcohol intoxication • Neurologic deficit
• Seizure
Coagulopathy • Warfarin use or coagulopathy (considered • Not part of the final rule (only • Suspected coagulopathy
a preexclusion criterion; i.e., not tested in one patient in the original (includes aspirin use)
original study) study met this criterion, pre-
venting evaluation)
Medium Risk for Brain Injury Detection by CT
(adding these criteria improves sensitivity but
reduces specificity)
• Amnesia before impact of ≥30 minutes
• Dangerous mechanism¶
Data from Haydel MJ, Preston CA, Mills TJ, et al: Indications for computed tomography in patients with minor head injury. N Engl J Med 343:
100-105, 2000; Mower WR, Hoffman JR, Herbert M, et al: Developing a decision instrument to guide computed tomographic imaging of blunt head
injury patients. J Trauma 59:954-959, 2005; Stiell IG, Wells GA, Vandemheen K, et al: The Canadian CT Head Rule for patients with minor head
injury. Lancet 357:1391-1396, 2001.
*For convenience of comparison, similar characteristics for the rules are paired. The order of variables is not important, as any single violation of a
rule requires CT imaging. Each bulleted item is a separate item requiring consideration.
†Exclusion criteria: No history of trauma, GCS <13, age <16 years, warfarin use or coagulopathy, obvious open skull fracture, unstable vital signs,
seizure before presentation.
‡Normal cranial nerves and normal strength and sensation in the arms and legs, as determined by a physician on the patient’s arrival at the emer-
gency department
§Including hemotympanum, raccoon eyes, cerebrospinal fluid, otorrhea or rhinorrhea, and Battle’s sign.
¶A pedestrian struck by a motor vehicle, an occupant ejected from a motor vehicle, or a fall from ≥3-foot elevation or five stairs.
was 12.6%, compared with 7.8% in those younger than injury on repeat head CT varied from 8% to 67%, with
65 years.178 The CCHR classifies patients older than 65 resulting neurosurgical intervention in 0% to 54% of
years as high risk and therefore in need of imaging in patients.206 The review’s authors cite a variety of expla-
the case of traumatic loss of consciousness. nations for this dramatic variability in study outcomes,
including selection bias, spectrum bias (studies with
more severely injured patients being more likely to
Is a Repeated Head CT Required for Patients With show unfavorable outcomes), and poor definitions of
Abnormal Head CT After Blunt Trauma? injury progression. Risk factors including coagulopathy,
A range of reported progression in CT findings has poor GCS score, and high overall injury severity appear
been published, with few clear-cut indications for to be associated with worsening CT abnormalities,
safely omitting repeat scan. A systematic review from but methodologic flaws in the studies reviewed make
2006 found that the range of reported progression of more specific recommendations impossible. Since the
Box 1-24: Prediction Rule for Head What Is the Best Imaging Modality for DAI?
Injury in Children 2 Years and Older As the name implies, DAI is damage to white matter
tracts throughout the brain, thought to occur as the
CT is not indicated for patients meeting these criteria. result of shearing from rapid deceleration, often with a
ll Normal mental status rotational component.209 There is some debate as to the
ll No loss of consciousness clinical scenarios in which this injury occurs, with some
ll No vomiting arguing that DAI is a feature only of severe injury while
ll Nonsevere injury mechanism
others suggesting it as a mechanism underlying postcon-
ll No signs of basilar skull fracture
cussive syndromes in patients with normal CT.210 Stud-
ll No severe headache
ies in patients with mild head injury are problematic,
From Kuppermann N, Holmes JF, Dayan PS, et al. Identifica- as it is unclear whether MR abnormalities are truly evi-
tion of children at very low risk of clinically-important brain dence of CT-negative DAI or, rather, false-positive MR
injuries after head trauma: A prospective cohort study. Lancet findings. CT is generally thought to be poor in detecting
374:1160-1170, 2009. Severe mechanism of injury: motor these changes, though a gold standard for comparison is
vehicle crash with patient ejection, death of another passenger,
often lacking or limited to comparison with MRI. A pro-
or rollover; pedestrian or bicyclist without helmet struck by a
motorized vehicle; falls of more than 1.5 m [5 feet]; or head spective study in 1988 compared CT and MR for identi-
struck by a high-impact object. fication of blunt traumatic head injuries, but advances in
both modalities have rendered its results invalid. A 1994
study found the modalities to be complementary for
head trauma, with MR substantially more sensitive for
publication of that review, a prospective study of level DAI.211 Subsequent studies have often compared new
I trauma center patients with an abnormal head CT has MR image sequences such as fluid attenuated inver-
addressed some of the issues raised by that review. The sion recovery and DWI to other MR sequences, using
study stratified patients according to GCS (mild: GCS = as a gold standard a clinical definition of DAI (loss of
13-15; moderate: GCS = 9-12; and severe: GCS <9) and consciousness persisting more than 6 hours after injury,
indication for repeat CT (routine vs. indicated by neu- no hemorrhage on CT) plus imaging criteria (presence
rologic deterioration). Among patients undergoing CT of white matter injury on MRI). This type of study, in
for neurologic deterioration, a medical or surgical inter- which the gold standard or reference used to determine
vention followed CT in 38%. In contrast, among patients the accuracy of the experimental test incorporates that
undergoing routine repeat CT, 1% underwent an inter- test, suffers from incorporation bias.212 The sensitivity of
vention—in both cases, in patients with a GCS score MRI may be overestimated as a result.
below 9. The authors conclude that repeat CT is war-
ranted in any patient with neurologic deterioration and
routine repeat CT may be warranted among patients
SUMMARY
with a GCS score below 9. No interventions occurred Emergency imaging of the brain is required for a num-
in patient with a GCS score of 9 or higher undergoing ber of presenting chief complaints. Clinical decision
routine head CT, but this study is too small to conclude rules can guide imaging in some instances, particu-
with certainty that routine repeat CT is never neces- larly trauma. Noncontrast CT is the most widely used
sary in this group.207 Other recent retrospective studies modality, but some processes, such as ischemic stroke,
also suggest that routine repeat head CT is not likely to vascular dissection, and SAH, cannot be ruled out by
change clinical management in the absence of a deterio- noncontrast CT alone. Emergency physicians should
rating neurologic examination, but a larger prospective understand when more diagnostic testing is required
study will be needed to more stringently define those after a normal CT. A systematic approach to CT inter-
patients with abnormal CT after blunt trauma in whom pretation can improve the accuracy of interpretation by
repeat CT can be deferred.208 emergency physicians.

CT Scan Head N Brain

Uploaded by

Copyright:

Available Formats

CT Scan Head N Brain

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

CT Scan Head N Brain

Uploaded by

Copyright:

Available Formats

CHAPTER 1 Imaging the Head and Brain

Joshua Broder, MD, FACEP, and Robert Preston, MD

interpretation of radiographic studies. A study simulat- by neuroradiologists.17,18 Undoubtedly, improvements

bone (≈+1000HU) bone (≈+1000HU)

air outside air outside

CSF (water) CSF (water)

Abnormal Asymmetry: Mass Effect

ethmoid air frontal

right maxillary left maxillary

right zygomatic air-fluid levels in

hemorrhage. Debate exists about the accuracy of CT in Subarachnoid Hemorrhage

Box 1-2: CT Findings of Subarachnoid

Box 1-3: CT Findings of Epidural

ll CT Appearance: Variable white to gray on

SDHs, or acute hematomas in anemic patients, may

Subdural Hematoma Indications for Surgery in Subdural Hematoma.

ventricles gland and choroid plexus calcifications also have stereo-

Indications for Surgery in Parenchymal Hemor-

Figure 1-18. Three variations of

Box 1-4: CT Findings of Subdural

ll Appearance: Variable white to gray on right lateral

vasogenic edema surrounding a lesion. Neoplasms often

normal posterior horn

surrounding region surrounding

TABLE 1-3. Early Ischemic CT Changes Within

normal MCA, hyperdense gray matter

Figure 1-27. Early ischemic hypo­

vasogenic edema both occur. Ion gradients run back

Hypodensity of Ischemic Stroke Versus Vasogenic

Are Ischemic Changes a Contraindication to t-PA?

white matter (myelinated axon tracts)

Box 1-7: Relationship Among Intra- Box 1-8: CT Findings of Cerebral

normal Figure 1-32. Normal cerebro­

Figure 1-33. Cerebral atrophy,

quadrigeminal plate enlarged

Figure 1-35. Cerebral edema, quadrigeminal

Figure 1-36. Comparison of cere­

the national average charge for head CT as $996 and for

Box 1-9: CT Findings of Acute Hydroc Box 1-10: CT Findings of Increased

From ACEP; Edlow JA, Panagos PD, Godwin SA et al. ­Clinical

contrast in stenotic region of

left MCA aneurysm

use of thrombolytics in patients presenting after an arbi-

However, routine use of CTPSs in the hospital setting

Ultrasonography for Ischemic Stroke Stroke Neuroimaging Summary

Box 1-13: ABCD(2) Score for Suspected Subarachnoid Hemorrhage

found that among all adult ED patients with new-

Box 1-16: Proposed Decision Rule Box 1-17: National Emergency

LOC, loss of consciousness.

emergency practice. In the United Kingdom, skull radio-

TABLE 1-8. Canadian Computed Tomography Head

From Mower WR, Hoffman JR, Herbert M, et al: Developing

You might also like

TABLE 1-3. Early Ischemic CT Changes Within

Figure 1-27. Early ischemic hypo

normal Figure 1-32. Normal cerebro

Figure 1-36. Comparison of cere

From ACEP; Edlow JA, Panagos PD, Godwin SA et al. Clinical

TABLE 1-8. Canadian Computed Tomography Head