Noncontrast Head CT

Noncontrast CT is the most commonly ordered head imaging test in the ED, used in up to 12% of all adult ED visits.^5,6 It provides information about hemorrhage, ischemic infarction, masses and mass effect, ventricular abnormalities such as hydrocephalus, cerebral edema, sinus abnormalities, and bone abnormalities such as fractures. Although dedicated facial CT provides more detail by acquiring thinner slices through the region of interest or by changing the patient’s position in the scanner during image acquisition, general information about the face and sinuses can be gleaned from a generic noncontrast head CT, as described in detail in Chapter 4 on facial imaging. The American College of Radiology recommends acquisition of contiguous or overlapping slices with slice thickness no greater than 5mm. For imaging of the cranial base, the ACR recommends the thinnest slices possible. If multiplanar or three-dimensional reconstruction is planned, slice thickness should be as thin as possible and no greater than 2-3 mm.^6a

Contrast-Enhanced Head Computed Tomography

Contrast-enhanced head CT is usually performed following a noncontrast CT, and the two are then compared. In a contrast-enhanced head CT, IV contrast is injected, usually through an upper extremity intravenous catheter. A time delay is introduced to allow the venous contrast to pass to the brain. Depending on the delay between contrast injection and image acquisition, the resulting CT may be a CT cerebral angiogram (CTA) or a CT cerebral venogram (CTV). The CT data may be reconstructed in any of several planes (typically axial, sagittal, or coronal) or even three-dimensionally. Cerebral perfusion can be mapped with IV contrast, using a technique described in more detail later. Contrast CT is useful for depicting abnormal vascular structures, such as aneurysms or arteriovenous malformations; for demonstrating abnormal failure of filling of vascular structures, such as sagittal sinus thrombosis (akin to demonstrating a filling defect in chest CT for pulmonary embolism); and for demonstrating neoplastic, inflammatory, and infectious processes. These latter abnormalities typically have increased regional blood flow, consequently receive abnormally high levels of vascular contrast agents, and may demonstrate ring enhancement, a circumferential increase in density around a lesion on CT occurring after IV contrast administration.

Sinus Trauma

Facial fractures are discussed in more detail in the dedicated chapter on facial imaging (Chapter 2). In trauma, fractures through the bony walls of sinuses result in bleeding into the sinus cavity. While the trauma patient remains in a supine position, this blood accumulates in the dependent portion of the sinus, forming an air–fluid level visible on CT. Previously existing sinus disease may be visible as circumferential sinus mucosal thickening, rather than as an air–fluid level. Inspect the sinuses carefully for air–fluid levels, as these may indicate occult fractures. In trauma, opacification of sinuses should be considered evidence of fracture until proven otherwise, as the fracture itself may be hard to identify (Figures 1-8 and 1-9). The ethmoid sinuses are small and may be completely opacified by blood in the event of fracture. Opacified ethmoid sinuses should increase suspicion of a medial orbital blowout fracture. Air–fluid levels in the maxillary sinus may be associated with inferior orbital blowout fractures, because the inferior wall of the orbit is the superior wall of the maxillary sinus. The frontal sinus is less easily fractured, as its anterior and posterior plates are thick and resistant to trauma. Fracture of the anterior wall of the frontal sinus is relatively less concerning, requiring plastic surgery or otolaryngology consultation. Fracture of the posterior wall of the frontal sinus is a potential neurosurgical emergency due to communication of the sinus space with the CSF. Look for intracranial air whenever frontal

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.)

Box 1-1 CT Findings of: Sinus and Mastoid Pathology

• Window setting: Bone windows helpful for smaller air-space evaluation.

• CT Appearance: Normally air filled (black), no air–fluid levels

• Significance

Air–fluid levels or opacification in the setting of trauma may indicate fracture.

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.

Mastoid opacification without trauma may indicate mastoiditis when clinical signs and symptoms are present.

Sinus Infections

In the absence of trauma, sinus mucosal thickening and air–fluid levels may be normal findings. They should not be used to make a diagnosis of bacterial sinusitis in the absence of strong clinical evidence, as they are nonspecific and may occur in allergic sinusitis or even asymptomatic patients. The mastoid air cells are not normally fluid filled, and in the presence of mastoid tenderness and erythema, their opacification on CT is evidence of mastoiditis (Figure 1-10). Sinus infections are discussed in more detail in Chapter 2, Imaging the Face.

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.

Subarachnoid Hemorrhage

SAH is blood within the subarachnoid space, which includes the sulci, Sylvian fissure, ventricles, and basilar cisterns surrounding the brainstem (Figure 1-14).

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.

Fresh SAH appears white, although the appearance varies depending on the ratio of blood to CSF.²⁴ CT is 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. Because subarachnoid blood may diffuse into adjacent regions, it may defy the guideline that hemorrhage and other abnormalities disturb normal brain symmetry. In other words, large amounts of SAH, including hemorrhage into cisterns, may actually result in a symmetrical-appearing head CT. Beware of this possibility when inspecting the brain for abnormalities. Familiarity 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 “star.” When filled with blood, it appears as a symmetrical white star (see Figure 1-14). The quadrigeminal plate cistern is usually visible on the same axial CT slice and has the appearance of a black “smile” when filled normally with CSF. When filled with subarachnoid blood, it becomes a white “smile.”

As time elapses from the moment of hemorrhage, blood will diffuse through the subarachnoid spaces, like a drop of food coloring dropped into a glass of water. Thus a bright white punctate finding on head CT is not

Box 1-2 CT Findings of Subarachnoid Hemorrhage

• Appearance: White on brain windows

• Location: Localized or diffuse

Sulci

Fissures

Ventricles

Cisterns

• Shape: Assumes shape of surroundings

• Pearl: Suspect increased ICP, look for signs of diffuse edema

• Pearl: Look for abnormal white “star” sign and “smile” sign from SAH in suprasellar and quadrigeminal plate cisterns

likely to be SAH, especially hours after the onset of clinical symptoms. Figure 1-14 shows several examples of SAH, involving different brain regions. SAH may be accompanied by other important changes, including hydrocephalus and cerebral edema, discussed later. Box 1-2 summarizes findings of SAH.

Epidural Hematoma

Epidural hematoma (EDH) is a collection of blood lying outside the dura mater, between the dura and the calvarium. It is almost always a traumatic injury, commonly resulting from injury to the middle meningeal artery. Because blood is extravasating from an artery under high pressure, rapid enlargement of the hematoma may occur, leading to significant mass effect, midline shift, and herniation. The common CT appearance is a biconvex disc or lens, collecting in the potential space between the calvarium and the dura mater.²⁸ This shape occurs because the more superficial aspect of the EDH conforms to the curve of the calvarium, while the inner aspect expands and presses into the dura. The dura is usually tethered to the calvarium 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. The usual location of an EDH is temporal, although EDHs occasional occur in other locations. Transfalcine herniation may occur with EDHs, so the midline of the brain should be carefully inspected on CT for midline shift or compression of the lateral ventricle. The swirl sign, described as a bright white vortex or “swirl” within the EDH, has long been considered a finding of active bleeding and should be interpreted as a sign of continued expansion, although recent studies have questioned the prognostic significance of this finding.^29-32 Figures 1-15 and 1-16 show several examples of EDHs, with the classic findings described earlier. Interestingly, the volume of hematoma has not been shown to correlate with preoperative neurologic status or 6-month postoperative status.³³ Box 1-3 summarizes findings of EDH.

Figure 1-15 Epidural hematoma (EDH), noncontrast CT, brain windows.

EDHs are frequently the result of traumatic injury to the middle meningeal artery at the site of a temporal bone fracture. They characteristically have a biconvex (lens) shape. Because they lie between the calvarium and the dura, they are restricted from extending beyond suture lines, where the dura is tightly adherent to the skull.

Sutures are not actually visible on this brain window setting but could be readily identified by use of bone windows. Suture locations are pointed out in 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 left side (cropped out of the field of image). Despite this, several classic features of EDH are visible:

• Lenslike or biconvex disc shape

• Temporal location, with associated depressed temporal bone fracture

• No crossing of suture lines

• Mass effect with midline shift

• Swirl sign (heterogeneous appearance suggesting active bleeding)

• Elevated intracranial pressure, with small ventricles and no visible sulci

(From Broder J, Preston R: An evidence-based approach to imaging of acute neurological conditions. Emerg Med Pract 9(12):7, 2007.)

Figure 1-16 Epidural hematoma (EDH), noncontrast CT, brain windows.

This EDH (arrowheads) demonstrates several classic features, including a lenslike or biconvex disc shape and a temporal location. Although it appears small on this slice, the entire CT may reveal a different story. On this slice, sulci are absent, raising concern for elevated intracranial pressure.

Subdural Hematoma

Subdural hematoma (SDH) (Figures 1-17 and 1-18) is a collection of blood between the dura mater and the brain surface. SDHs usually occur from traumatic injury to bridging dural veins, although a history of trauma is not always found. SDHs may be self-limited in size due to the lower pressure of venous bleeding, but they can become enormous, causing significant mass effect, midline shift, and herniation. They may also rebleed after an initial delay, resulting in expansion. Moreover, they are frequently markers of significant head trauma, and patient outcomes may be compromised by associated diffuse axonal injury (DAI) (described later) or edema. The typical CT appearance of an SDH is a crescent, with the convex side facing the calvarium and the concave surface abutting the brain surface. The shape of SDHs results from their accumulation between the dura and the brain surface. Because they lie between the dura and the brain, they are not restricted by attachment sites between the dura and the calvarium at sutures. Consequently, SDHs may cross suture lines. Moreover, each cerebral hemisphere is wrapped in its own dura, so SDHs typically do not cross the midline but instead may continue to follow the brain surface into the interhemispheric fissure.

Figure 1-17 Subdural hematoma (SDH), noncontrast CT, brain windows.

SDHs are most often the result of shearing of dural veins from blunt trauma. They usually have a crescent shape and may cross 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 of the dura to the calvarium at sutures. In this figure, the sutures themselves are not visible due to the window setting, but their locations are labeled to illustrate this point.

This SDH demonstrates several classic features:

• Crescent shape

• Crossing of suture lines

• Mass effect with midline shift

• Elevated intracranial pressure (ICP), with small ventricles

• No visible sulci (due to effacement from elevated ICP)

(From Broder J, Preston R: An evidence-based approach to imaging of acute neurological conditions. Emerg Med Pract 9(12):5, 2007.)

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.

The color may vary depending on the age of the SDH (see Figure 1-18). Fresh subdurals are typically brighter white (or lighter gray) than the adjacent brain. Older

Box 1-3 CT Findings of Epidural Hematoma

• CT Appearance: Variable white to gray on brain windows

• Location: peripheral to brain, variable but usually temporal region

• Shape: Biconvex disc or lens

• Pearl: Does not cross suture lines

• White swirl sign means active bleeding

• Significance: May cause mass effect and herniation

Look for midline shift

Look for effacement of ventricles and sulci

• Surgical indications³⁴: 15-mm thickness or 5-mm midline shift

SDHs, or acute hematomas in anemic patients, may become similar in density (isodense) to the adjacent brain and thus may be difficult to detect.^35,36 Clues to their presence include the obliteration of sulci on the brain surface and mass effect resulting from the SDH. Still older SDHs may become similar in density or color to the CSF surrounding the brain and thus may be difficult to recognize. Sometimes SDHs are multicolored or layered, indicating blood of varying ages. Box 1-4 summarizes findings of SDH.

Intraparenchymal Hemorrhage

Intraparenchymal hemorrhage (Figures 1-19 and 1-20), or hemorrhage within the substance of the brain matter, may occur in trauma or spontaneously, perhaps as a complication of hypertension. The appearance is generally bright white acutely. The size may vary from punctate to catastrophically large, with associated mass effect and midline shift. For intraparenchmyal hemorrhage, mass effect such as midline shift or ventricular effacement should be assessed. Signs of increased ICP should be identified. Particularly for smaller punctate hemorrhages, care must be taken not to mistake hemorrhage for normal benign calcifications of the pineal gland, choroid plexus, and meninges, or vice versa. Calcifications can usually be distinguished from hemorrhage as the former remain bright white on bone windows. In addition, if the density is measured with PACS tools, calcifications have very high density, near +1000HU, while hemorrhage has a density around +40 to +70HU. Pineal gland and choroid plexus calcifications also have stereotypical locations (Figure 1-21) that aid in their recognition. Calcifications are shown in Figure 1-21.

Figure 1-19 Intraparenchymal hemorrhage, noncontrast CT, brain windows.

Acute intraparenchymal hemorrhage appears white on CT brain windows. Hemorrhage in the patient’s left frontal region is creating mass effect with midline shift. The left lateral ventricle has been completely effaced. The single visible ventricle is the right lateral ventricle.

A calcified mass in the right occipital region must be differentiated from acute hemorrhage. Calcifications are extremely bright white on brain windows—as white and dense as bone. On bone windows, they remain white, while hemorrhage does not. The density can also be measured using PACS tools. Bone has a density near +1000HU, while blood 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.)

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).

Figure 1-21 Calcifications, noncontrast CT, brain windows.

Calcification of the choroid plexi is a frequent incidental finding that may resemble punctate intraparenchymal hemorrhage. Clues are bright white density (equal to that of bone), location in the posterior horns of the lateral ventricles, and frequent bilaterality.

This patient also has a calcified meningioma. Meningiomas are common benign neoplasms that may become quite large. A well-circumscribed, rounded appearance and calcification are common.

(From Broder J, Preston R: An evidence-based approach to imaging of acute neurological conditions. Emerg Med Pract 9(12):8, 2007.)

Masses

Masses are best delineated on CT with IV contrast or on MRI. However, noncontrast CT can demonstrate a variety of masses if they are of sufficient size. In some cases, the mass itself may not be seen, but secondary findings such as mass effect, calcification (see Figure 1-21), or vasogenic edema (Figure 1-22) (see the discussion in the next section) may occur. IV contrast is useful in detecting masses because they are generally extremely vascular and thus enhance in the presence of contrast material. On noncontrast CT, masses may be denser than surrounding brain if they are calcified. Examples are meningiomas, which are often seen as midline structures emanating from the falx cerebri.

Figure 1-22 Masses, noncontrast CT, brain windows.

A mass with surrounding vasogenic edema, which has a hypodense (dark gray) appearance. Neoplasms frequently are associated with vasogenic edema, named for the putative cause, which is abnormal blood vessels that allow extravasation of fluid.

This form of edema appears hypodense, like an ischemic infarct, but is not restricted to a vascular territory. An abscess might appear similar.

Masses often exert mass effect. Here, the edema surrounding the mass is compressing the posterior horn of the left lateral ventricle. Notice also the effacement of the sulci in the region of the mass.

(From Broder J, Preston R: An evidence-based approach to imaging of acute neurological conditions. Emerg Med Pract 9(12):8, 2007.)

Vasogenic Edema

Malignant primary brain neoplasms or metastatic lesions often appear hypodense (darker or blacker) compared with normal brain. This appearance is typical of localized

Box 1-4 CT Findings of Subdural Hematoma

• Appearance: Variable white to gray on brain windows

• Location: peripheral to brain or interhemispheric; variable location around brain perimeter

• Shape: Crescent

• Pearl: May cross suture lines

• Significance: May cause mass effect and herniation

Look for midline shift

Associated with other brain injuries, look for effacement of ventricles and sulci

• Surgical indications³⁷: 10-mm thickness or 5-mm midline shift

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 endothelial junctions, allowing fluid to extravasate into the interstitium, which causes vasogenic edema. This increased fluid content reduces the density of brain tissues toward that of water (zero on the Hounsfield scale), resulting in a hypodense appearance on CT. In addition, the mass itself and associated local edema increase the volume of brain tissue, resulting in local mass effect, including the effacement of ventricles (see Figure 1-22, effacement of the posterior horn of the lateral ventricle) and effacement of sulci as adjacent gyri expand in size (see Figure 1-22).

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 territory within the brain, whereas hypodensity associated with ischemic stroke does. Vasogenic edema responds to treatment with dexamethasone, whereas steroids are not indicated for other forms of cerebral edema such as traumatic edema do not. In fact, a multicenter randomized controlled trial of corticosteroid therapy in patients with traumatic brain injury showed an increased risk of death and severe disability from steroid use.³⁸

As described earlier, midline shift associated with a mass should be carefully assessed during inspection of the brain on brain windows.

Abscesses

Abscesses may be visible on noncontrast CT as hypodense regions (Figure 1-23), occasionally with air within them. This appearance may be nonspecific, and a differential diagnosis including toxoplasmosis, mass with vasogenic edema, or central nervous system lymphoma should be considered, depending on the patient’s clinical scenario. Abscesses, toxoplasmosis, neurocysticercosis (Figure 1-24), and masses all may undergo ring enhancement, an increase in density around a lesion after administration of IV contrast (Box 1-5). This reflects increased blood flow in the vicinity of the lesion, as well as leaky vascular structures that allow extravasation of contrast in the region.

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.

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.

Ischemic Stroke and Infarction

Ischemic stroke accounts for 85% of strokes.³⁹ It is potentially one of the most important indications for head CT and is an area in which the interpretation of CT by emergency physicians might play the greatest role by shortening the time to diagnosis. One obvious reason is the 3-hour or 4.5-hour window for administration of IV t-PA—an intervention that is still fiercely debated in the emergency medicine community and that has been reviewed elsewhere.⁴⁰

Understanding CT findings of acute ischemic stroke is important—for those who do not believe in

Box 1-5 CT Findings of Enhancement

• Increased brightness (Hounsfield units) after IV contrast administration

• Seen in

Infection

Inflammation

Neoplasm

Vascular lesions

administration of t-PA, they provide yet another argument against the treatment, while for those who would use t-PA in select patients, they may allow more rational and safer patient selection. Apart from t-PA administration, rapid diagnosis of ischemic stroke may allow the emergency physician to make better-informed decisions about patient management and disposition. If new stroke therapies such as intra arterial thrombolysis and clot retrieval become widely accepted and available, rapid CT interpretation for ischemic stroke may become even more valuable. Interventional radiologic therapies for ischemic stroke are reviewed in Chapter 16.

A complex cascade of events occurs to cause the evolving appearance of ischemic stroke on head CT. Initially, at the moment of onset of cerebral ischemia, no abnormalities may be seen on head CT—thus, this is one of the most difficult diagnoses for the emergency physician, as a normal CT may correlate with significant pathology. Studies have shown emergency physicians to be relatively poor at recognizing early ischemic changes, which we review here (Table 1-3 and Box 1-6).

TABLE 1-3 Early Ischemic CT Changes Within 3 Hours of symptom onset, Possibly Altering Management

How Early Does the Noncontrast Head CT Indicate Ischemic Stroke?

Analysis of the National Institute of Neurological Disorders and Stroke (NINDS) data shows that early ischemic changes are quite common in ischemic stroke, occurring in 31% of patients within 3 hours of stroke onset,⁴¹ in contrast to the widely held belief that ischemic strokes become visible on CT only after 6 hours.⁴² Some findings may occur immediately, such as the hyperdense middle cerebral artery (MCA) sign, while other findings may require time to elapse, with the gradual failure of adenosine triphosphate (ATP)–dependent ion pumps and resulting fluid shifts.

Hyperdense Middle Cerebral Artery Sign

The hyperdense MCA sign is a finding of hyperacute stroke, indicating thrombotic occlusion of the proximal MCA. This may be present on the initial noncontrast head CT immediately following symptom onset, since the finding does not require the failure of ion pumps and fluid shifts that lead to other ischemic changes on head CT. Because

Box 1-6 Early CT Findings of Acute Ischemia

From Patel SC, Levine SR, Tilley BC, et al: Lack of clinical significance of early ischemic changes on computed tomography in acute stroke. JAMA 286:2830-2838, 2001.

• Hyperdense MCA sign

• Loss of gray–white differentiation

Insular ribbon sign

Cortical sulcal effacement

• Ischemic focal hypoattenuation

MCA territory

Basal ganglia

this lesion leads to ischemia in the entire MCA territory, typically the patient with this finding will have profound hemiparesis or hemiplegia on the contralateral side, as well as other findings such as language impairment, depending on the side of the lesion. In other words, this finding is not associated with mild or subtle strokes. The presence of a hyperdense MCA sign is an independent predictor of neurologic deterioration.⁴³ However,the dense MCA sign is only visible in 30% to 40% of patients with stroke affecting the MCA territory.^71,72 It may seem surprising that this vascular abnormality is visible on noncontrast head CT. As the name implies, the MCA appears hyperdense (bright white) compared with the normal side. A specific Hounsfield unit threshold of greater than 43 units has been recommended to avoid false positives.⁴⁴

Use your knowledge of the location of the patient’s neurologic deficits to direct you to the likely side of the lesion, which will be on the contralateral side. Then use the normal symmetry of the brain to help you identify this abnormality. A related finding, the MCA “dot” sign, has been validated by angiography and found to be a specific marker of branch occlusion of the MCA. This sign appears as a bright white dot in the sylvian fissure on the affected side.⁴⁵ Figure 1-25 shows the hyperdense MCA sign.

Figure 1-25 Hyperdense middle cerebral artery (MCA) sign, noncontrast CT, brain windows.

The hyperdense MCA sign is a CT finding of thrombosis of the MCA. Importantly, this is a sign seen on noncontrast CT—the typical initial imaging study obtained in patients with suspected acute stroke. It can be seen in the immediate hyperacute stages of thrombotic–ischemic stroke and may guide therapy, such as intra arterial tissue plasminogen activator administration. On CT, the 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, the normal MCA is not visible while the left MCA is thrombosed and demonstrates the hyperdense MCA sign.

(From Broder J, Preston R: An evidence-based approach to imaging of acute neurological conditions. Emerg Med Pract 9(12):12, 2007.)

Gray–White Differentiation

Understanding this finding of stroke requires a brief and simple review of neuroanatomy. Gray matter is brain tissue without myelin—examples include the cerebral cortex, lentiform nucleus, caudate, and thalamus. White matter is composed of myelinated axons in brain tissue—rendered white on gross pathologic section by the high lipid content of the myelin sheath. Recall from our earlier discussion of Hounsfield units that lower density on CT means a darker color—low-density fat appears a darker gray than does higher-density water. Thus, the higher fat content of white matter makes it appear darker on CT. In other words, on a normal head CT, gray matter is whiter and white matter is grayer. Figure 1-26 shows the normal gray–white matter boundary.

Figure 1-26 Normal gray–white matter differentiation, noncontrast CT, brain windows.

Myelinated regions (white matter) have a greater fat content than unmyelinated regions (gray matter). As a consequence, and perhaps counterintuitively, white matter has lower density and appears darker than gray matter on computed tomography. The dotted line on the patient’s right outlines the border between gray and white matter. Trace this interface yourself on the patient’s left.

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 working. Ions equilibrate across membranes, and fluid shifts occur. Gray matter gains fluid, lowering its density, and as it does, its density becomes more similar to that of white matter. White matter also gains fluid, increasing its density slightly. Since differences in density are the reason that these tissues look different on CT, as their densities converge, their appearances become more similar, and it becomes more difficult to discern where gray matter ends and white matter begins. This change is called loss of gray–white matter differentiation, and it is an early finding of ischemic stroke, occurring within 3 hours after onset of ischemia.⁴¹ Figures 1-27 and 1-28 show an abnormal gray–white matter boundary.

Figure 1-27 Early ischemic hypodensity, 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 hypoattenuation predict an increased risk of hemorrhage and are relative contraindications for tissue plasminogen activator.

Figure 1-28 Gray and white matter differentiation, noncontrast CT, brain windows.

When ischemia renders the gray–white interface less discrete, the computed tomography appearance is called loss of gray–white differentiation. In this example, the differentiation is normal on 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 the frank hypodensity of ischemic stroke.

Hypodensity in Ischemic Stroke

Ischemic brain looks hypodense, or darker than normal brain in the same anatomic region. This change occurs for the same general reasons as does loss of gray–white differentiation. As neurons deplete stores of ATP, cytotoxic and vasogenic edema both occur. Ion gradients run back toward equilibrium, and water shifts into gray matter, 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.

Figure 1-29 Early ischemic changes, noncontrast CT, brain windows.

Early ischemic changes may be visible within 3 hours of onset of ischemic stroke. They include sulcal effacement, loss of gray–white matter differentiation, and the insular ribbon sign. Sulcal effacement occurs as local edema develops, swelling brain matter and displacing the cerebrospinal fluid that normally fills sulci as the adjacent gyri become edematous. Loss of gray–white matter differentiation occurs as ion pumps fail, leading to equilibration of diffusion gradients and shift of fluid. The normal ability of computed tomography (CT) to differentiate gray from 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 region of a developing infarct renders the region darker on CT because of the presence of increasing amounts of fluid. This masks the normal differentiation between white and gray matter. The insular ribbon sign is another manifestation of this loss of gray–white matter differentiation. The insula is a region of gray matter lining the lateral sulcus, in which 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 hypodensity of ischemic stroke is also becoming visible. Compare these to similar regions on the patient’s right side, where normal sulci and normal gray–white matter differentiation are seen.

(From Broder J, Preston R: An evidence-based approach to imaging of acute neurological conditions. Emerg Med Pract 9(12):12, 2007.)

Figure 1-30 Early ischemic changes, noncontrast CT, brain windows.

(Same image as Figure 1-29.) Image contrast has been increased to accentuate gray–white matter differentiation. On a PACS image, you can adjust the window level and contrast. In a normal brain, you should be able to discern several white and gray matter structures.

Normal gray–white matter differentiation is subtle. Gray matter has lower lipid content than myelinated white matter and therefore appears brighter on computed tomography (CT). On CT, this leads counterintuitively to gray matter appearing whiter and white matter appearing grayer.

Normal gray matter areas include the cerebral cortex, lentiform nucleus, caudate, and thalamus. White matter tracks, including the internal capsule, separate these structures.

(From Broder J, Preston R: An evidence-based approach to imaging of acute neurological conditions. Emerg Med Pract 9(12):12, 2007.)

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.)

Are Ischemic Changes a Contraindication to t-PA?

Early ischemic stroke findings were not used as exclusion criteria in the NINDS trial, which required only the absence of hemorrhage on initial head CT.⁴⁶ However, multiple studies following NINDS have shown an increased risk of intracranial hemorrhage, bad neurologic outcomes, and death in patients with early ischemic changes on head CT.^47,48 Ischemic changes are relative contraindications to t-PA administration, and their presence may suggest that greater than 3 hours have elapsed from symptom onset, in which case systemic t-PA may be contraindicated. In addition, the Food and Drug Administration, American Heart Association, and American Academy of Neurology specifically recommend against administering t-PA if early signs of major infarction are present, because of increased risk of intracranial hemorrhage.^49-51 MCA infarction greater than one third of the MCA territory predicts increased bleeding risk if t-PA is given, and it was poorly detected by radiologists, neurologists, and emergency physicians in past studies.^18,52,53 In addition, the greater the extent of ischemic changes on CT, the higher the risk of bleeding, as demonstrated in the second multinational European Cooperative Acute Stroke Study (ECASS-II).⁴⁸

The many noncontrast CT findings of ischemic stroke may seem too much to hope to remember, and their clinical relevance may appear unclear. A few simple rules can make sense of this. First, a normal head CT is perhaps the most likely finding if the patient presents within 3 hours of symptom onset. In this setting, the most important job of the emergency physician in interpreting the head CT is to rule out hemorrhage. Second, in the presence of significant unilateral neurologic abnormalities, the hyperdense MCA sign should be sought. Third, early changes such as loss of gray–white differentiation and hypodensity should be identified, again using the patient’s clinical symptoms to direct you to the likely abnormal side of the brain. These early changes may imply either an earlier time of onset than suggested by the history or a massive stroke in progress.

Cerebral Edema

Diffuse cerebral edema is an abnormality of the brain, but it is manifest through compression of CSF spaces. It can logically be evaluated under B or C in our mnemonic. Cerebral edema can result from many pathologic processes, including trauma, anoxic brain injury, carbon monoxide poisoning, and systemic fluid and electrolyte abnormalities. The appearance is therefore not diagnostic of the underlying etiology. As the brain swells, several visible changes occur on noncontrast CT. CSF spaces become collapsed as they give way to the increasing volume of solid brain tissue. As a result, the lateral ventricles become slitlike and ultimately become obliterated. In addition, the sulci become effaced as the gyri swell. The normal rim of CSF surrounding the brain disappears. The cisterns surrounding the brainstem become compressed, and risk of herniation rises. Moreover, as the ICP rises, the cerebral perfusion pressure falls

(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 edema. Box 1-8 summarizes findings of cerebral edema.

Diffuse Axonal Injury

DAI is the widespread shearing of long axons that occurs as the result of deceleration injury. Common clinical scenarios include high-speed motor vehicle collisions and falls from great height. This injury is not typical of blows to the head or penetrating brain injury. The CT appearance is nonspecific: normal in the hyperacute phase, often followed by cerebral edema over hours to days. Punctate intraparenchymal hemorrhage may occur as well. Often, other traumatic brain injury will be evident, such as SDH or EDH.

The prognosis is poor, and resolution of CT findings may not equate with clinical improvement. MRI is thought to be more diagnostic.⁵⁴

Box 1-8 CT Findings of Cerebral Edema

• Loss of CSF-containing spaces

Sulci effacement

Ventricular effacement

Cistern effacement

• Loss of gray–white differentiation

• Significance

Increased ICP

Global brain ischemia from decreased cerebral perfusion pressure

Normal Findings That May Simulate Disease

Several common incidental findings may simulate disease. These include calcifications in the choroid plexus of the posterior horns of the lateral ventricles (recall that the choroid plexus secretes CSF) (see Figure 1-21) and calcifications in the pineal gland.⁵⁵ These should not be confused with hemorrhage, because they have a greater density (brighter white appearance) and a stereo typical location. The significance of these findings is unknown, although choroid calcifications have been associated with hallucinations in schizophrenia.⁵⁶

Why Can Two Patients With the Same CT Findings Have Markedly Different Neurologic Examinations?

Remember that CT offers a macroscopic snapshot in time of complex pathologic changes. It may be that a patient with severe neurologic impairment but a relative benign–looking head CT will soon develop changes such as cerebral edema due to neuronal injury that has already occurred or is ongoing. In other cases, a patient with a large SDH may appear surprisingly neurologically intact, whereas another patient with similar head CT findings is severely impaired. One explanation is the degree of DAI that may accompany abnormalities such as SDH. The patient with minimal deficits may have no DAI, whereas the patient with severe deficits may have severe DAI, which is not as evident on CT. Another of many factors that may determine clinical status is the amount of cerebral atrophy present before the injury. Atrophy is loss of brain volume and compensatory increase in the size of CSF-containing spaces, such as ventricles, cisterns, and sulci. When an injury occurs and cerebral edema or a space-occupying lesion such as an SDH develops, the presence of atrophy (see Figure 1-33) may be protective by allowing room for expansion of the pathologic lesion without leading to herniation or precipitous rises in ICP.

Sulci

Sulci, the CSF spaces between the undulating gyri of the brain surface, should appear black on brain windows. Normal sulci are visible but not prominent, and a thin ribbon of CSF should outline the entire brain. In cases of cerebral edema, the sulci may be completely effaced as the brain swells. In hydrocephalus, the volume of brain remains fixed but ventricles increase in size, leading to compression of the sulci. In cases of cerebral atrophy, loss of brain tissue volume leads to a compensatory increase in the size of sulci (Figures 1-32 through 1-36).

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.)

Ventricles and Hydrocephalus

Hydrocephalus is an important finding for emergency physicians because of its potential as a neurosurgical emergency. Untreated, hydrocephalus can result in tonsillar herniation, brainstem compression, and respiratory arrest.⁵⁷ In general, as hydrocephalus becomes severe, the lateral ventricles become significantly enlarged. Because the volume of the calvarium is fixed, and solid brain tissue is essentially incompressible, as the ventricles expand, other CSF spaces become compressed—consequently, the sulci become effaced. In contrast, in the patient with atrophy, the ventricles may appear dilated but sulci appear similarly enlarged. In communicating hydrocephalus, the axial CT at the level of the lateral ventricles and third ventricle can resemble the face of a Halloween pumpkin. The enlarged anterior horns of the lateral ventricles are the rounded eyes, and the normally slitlike midline third ventricle dilates to a rounded nose. The posterior horns of the two lateral ventricles resemble corners of a grin, though they do not connect in the midline. Depending on the location of obstruction to CSF flow, the fourth ventricle just posterior to the quadrigeminal plate cistern may become dilated, producing an O-shaped mouth. The quadrigeminal plate cistern itself may be effaced if significant downward herniation is occurring. This appearance has also been described as a cartwheel, with 5 spokes (the 2 anterior and 2 posterior horns of the lateral ventricles, plus the fourth ventricle) radiating from an axle (the third ventricle). In noncommunicating hydrocephalus, the obstruction may lie proximal to the fourth ventricle, which will then not appear dilated. Comparison with a prior head CT is always valuable in assessing for hydrocephalus, because ventricular size alone is a relatively poor predictor of ICP.⁵⁸ Normal ventricular size does not completely exclude the possibility of increased intracranial pressure.

A variety of CT criteria for acute and chronic hydrocephalus have been described (Box 1-9). Figure 1-34 shows CT findings of hydrocephalus. Figure 1-37 shows a disconnected ventriculoperitoneal shunt, which can result in hydrocephalus.

Atrophy

In cerebral atrophy, loss of brain volume results in relatively symmetrical increase in size of sulci and ventricles. The basilar cisterns should also remain patent (see Figure 1-33).

Computed Tomography Findings of Elevated Intracranial Pressure

A variety of CT findings may indicate elevated ICP, though none are completely predictive. Findings that suggest increased ICP are decreased ventricle size, decreased basilar cistern size, effacement of sulci, degree of transfalcine herniation (midline shift), and loss of gray–white matter differentiation (Box 1-10).⁵⁹

What CT Findings Are Contraindications to Lumbar Puncture?

Before performing lumbar puncture, confirm that findings of midline shift or elevated ICP are not present. The midline should be midline, the sulci should be evident, and the cisterns should be open. Ventricle size should not be excessive, particularly when compared to sulci. The evidence basis for assessment of ICP using CT is reviewed later in this chapter.

Computed Tomography for Stroke

CT is the most widely available, immediate imaging technique for patients presenting to the ED with signs and symptoms of stroke. Noncontrast head CT is rapid, taking less than 5 seconds for image acquisition using some 64-slice scanners.²² It is sensitive for detecting intracranial hemorrhage,²⁶ and immediate imaging is more cost effective than either delayed or selective imaging strategies.⁶⁹ Limitations do exist, however. Surrounding bone can obscure evidence of ischemic stroke, an artifact effect known as beam hardening. This problem can be minimized by requesting fine thickness cuts (~1 mm), but the risk of false negatives for stroke detection still exists—particularly when a vertebral–basilar distribution is present, because beam hardening is worsened by thick bone surrounding the posterior fossa.⁷⁰

Noncontrast CT findings of ischemic stroke were reviewed earlier in the section on head CT interpretation. The prognostic importance of these findings and implications for t-PA therapy are also discussed there. Although the sensitivity of noncontrast head CT for ischemic stroke increases beyond 24 hours, the sensitivity for ischemia-induced changes in the first six hours is relatively low at 66%. Thus, a normal CT is consistent with the diagnosis of acute ischemic stroke in a patient presenting with suggestive signs and symptoms.⁴⁷

For the emergency physician, several points about early ischemic changes should be emphasized. First, in contrast to the assertion in some emergency medicine texts that ischemic stroke becomes visible on noncontrast head CT only after 6 hours,⁴² the NINDS trial upon which t-PA therapy is largely based demonstrated that 31% of patients had early findings of ischemia within 3 hours of symptom onset (see Table 1-3).⁴¹ Although early ischemic changes were not an exclusion criterion for t-PA in the original NINDS trial,⁴⁶ subsequent research has shown a heightened risk of hemorrhagic conversion of ischemic stroke, poor neurologic outcomes, and death in patients with these changes.^47,48 As a consequence, and as mentioned earlier, the Food and Drug Administration, American Heart Association, and American Academy of Neurology recommend against use of t-PA in the patients with major early ischemic changes.⁷³ Specifically, early ischemic changes occupying an area one third the size of the MCA territory or one third of a cerebral hemisphere, cerebral edema, and midline shift are considered relative contraindications to t-PA because of the increased risk for hemorrhage.⁷⁴ When discussing the head CT findings with a radiologist before administration of t-PA, it is important to ask specifically about the presence and extent of these changes, in addition to asking about hemorrhage.

Standard contrast-enhanced head CT is rarely used for evaluation of stroke since it provides little additional information compared with noncontrast head CT. With the advent of multidetector CT scanners and spiral CT technology, however, CT angiography (CTA) can be performed to obtain images of the extracranial and intracranial vasculature from the aortic arch to the cranial vertex (Figs. 1-38 through 1-42). Images are acquired by administering a rapid bolus of IV contrast immediately after standard noncontrast head CT. The raw images can be acquired in as little as 60 seconds, and three-dimensional computer reconstructions can be performed in less than 15 minutes. The results might profoundly alter the course of management, because large artery occlusions correlate with National Institutes of Health Stroke Scale scores⁷⁵ and may indicate a need for endovascular intervention (Discussed in Chapter 16). Generally, agreement between CTA and catheter angiography—still the gold standard for diagnosis of vessel stenosis—approaches 95%. For severe carotid artery stenosis, sensitivity for CTA approaches 100%, whereas the sensitivity for diminished flow in the circle of Willis is 89%.⁷⁶ Although traditional angiography may have subtle, additional benefits related to characterization of the plaque lesion, the noninvasive and rapid nature of CTA renders it an attractive option to the emergency physician, assuming that the risks of exposure to contrast and additional radiation are acceptable to the patient.

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).

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.)

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.

Figure 1-41 Computed tomography (CT) perfusion scan.

A CT perfusion scan provides a map of blood flow in ischemic stroke, potentially identifying ischemic areas that could be salvaged by reperfusion. Here, 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 setting but is not a routine part of most clinical practice. Erroneously programmed CT perfusion scans have been the cause of iatrogenic radiation injury to patients undergoing stroke evaluations.

(From Broder J, Preston R: An evidence-based approach to imaging of acute neurological conditions. Emerg Med Pract 9(12):13, 2007.)

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.)

CTA uses enhancement of the cerebral vasculature as a surrogate for estimating perfusion of the parenchyma. Targeted CT perfusion studies (CTPSs) can be performed simultaneously using the same bolus of contrast⁷⁷ and have a sensitivity and specificity for detecting ischemia of 95% and 100%, respectively.⁷⁸ By measuring the rise and fall in concentration of injected contrast over time, CTPSs are capable of even more direct estimates of cerebral perfusion than CTA, including measurements of cerebral blood volume and cerebral blood flow (see Figure 1-41). Quantifying these variables may allow clinicians to identify areas of the brain that, although ischemic, are potentially still viable—the so-called ischemic penumbra. This has implications for the clinician attempting to weigh the benefits of administering IV or intraarterial thrombolytic drugs against the risk of intracranial hemorrhage. Routine use of CTPSs could potentially allow a more precise prediction of outcome⁷⁹ and could even herald a paradigm shift in one of the indications for thrombolytic administration: rather than excluding the use of thrombolytics in patients presenting after an arbitrary time interval (e.g., 3 hours), thrombolytic therapy could be initiated or excluded based on actual visualization 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 contrast, so ischemia located outside the scanning level of interest can be missed,⁸¹ although this is partially alleviated by the use of multislice scanners or repeated contrast boluses. Third, despite the theoretical appeal, only small studies in limited populations exist that confirm the ability of CTPSs to detect infarct,⁸² to predict infarct location⁸³ and size,⁸⁴ and to predict final outcomes.^82,84 We await large study confirmation of these findings before routinely recommending their use in the ED setting. Because CT perfusion studies require that the same regions of the brain be repeatedly scanned over a period of a few minutes to acquire data about perfusion over time, focal areas of the brain receive higher levels of radiation exposure. Widely publicized accounts exist of patients receiving dangerous radiation exposures when CT scanners were misprogrammed, prompting lawsuits and an investigation by the US Food and Drug Administration.^84a

Magnetic Resonance Imaging for Acute Stroke

Standard MRI for stroke includes scout images, T1- and T2-weighted images and MRA. Increasingly available new-generation scanners incorporate additional high-sensitivity methods such as diffusion-weighted imaging (DWI), gradient echo pulse sequencing (GEPS), and perfusion-weighted imaging (PWI).

Obtaining DWI has been possible since 1985.⁸⁵ In brief, the technique involves detecting and processing a signal in response to the movement of water molecules caused by two pulses of radiofrequency. Ischemic changes can be detected in as little as 3 to 30 minutes after insult, and in a small study of 22 patients who presented within 6 hours of symptom onset, DWI was found to be 100% sensitive and specific.⁸⁶ In a subsequent study, DWI was found to have a far superior sensitivity compared to CT (91% vs. 61%).⁸⁷ When MRA is done simultaneous with DWI as part of a fast protocol to detect vascular stenosis, their combined use within 24 hours of hospitalization substantially improved the early diagnostic accuracy of ischemic stroke subtypes.⁸⁸ DWI may also be useful when the clinician encounters a patient with remote-onset neurologic defects. In patients presenting with a median delay of 17 days after symptom onset, clinicians gained additional clinical information one third of the time (including clarification of the vascular territory affected) by performing DWI in addition to conventional T2-weighted images. Of this third, the information was designated as “highly likely” to affect management strategy in 38%.⁸⁹

MRI is superior to CT at both detecting acute ischemic change^85,87 and visualizing the posterior fossa.^70,80 However, MRI has failed to supplant CT as the imaging modality of choice for stroke in the ED due to cost, availability of the requisite personnel, time, and a long-standing belief that MRI is not reliable for detecting intracerebral hemorrhage. At least the last two factors are being surmounted. New scanners are faster, with acquisition times in the range of 3 to 5 minutes, compared to 15 to 20 minutes previously.⁹⁰ With respect to hemorrhage, DWI has proved sufficient to exclude intracerebral hemorrhage,⁹¹ and in studies comparing GEPS with CT, GEPS was at least as useful for detecting acute intracranial hemorrhage and actually better at elucidating chronic hemorrhagic changes,^92-94 with sensitivity approaching 100% when interpreted by trained personnel.⁹² The issues of cost and personnel are more complex, however. MRI hardware costs and costs associated with imaging are roughly double those of CT.⁹⁵ Whether these costs will fall in the future or can be justified in the form of better outcomes, shorter hospital stays, or other measurable end points is unknown. Personnel issues are related not only to sheer manpower but also to qualitative training demands. MRI requires specially trained technicians spending more time per study compared to CT, and expert-level radiologists with extensive training in MRI interpretation must be employed, because interpretation is still not reproducible (though advocates of DWI point out that there is virtually no intra-observer or interobserver variability with this modality).⁸⁷

Similar to the ability of CTPSs to identify the ischemic penumbra, the combination of DWI and PWI makes it possible to make inferences about ischemia before permanent injury (infarction) has occurred. DWI provides a map of brain tissue that is ischemic and at high risk for infarction. PWI provides a map of brain tissue that is threatened by decreased blood flow but not yet demonstrating cellular injury marked by changes in diffusion. Typically, the hypoperfused region (PWI defect) is larger than the area of diffusion abnormality (DWI defect) early in stroke. The DWI–PWI mismatch is thought to represent the ischemic penumbra, a region that is potentially salvageable with aggressive reperfusion (either by thrombolytic therapy or by catheter-based mechanical interventions).⁹⁶

PWI is performed with standard MRI and MRA using gadolinium and requires a total imaging time of less than 15 minutes. PWI can be performed in cases of contraindication to gadolinium (a rare event, as gadolinium has been found safe in most instances, though recent fatal nephrogenic systemic fibrosis has been noted in patients with advanced renal disease; see Chapter 15⁹⁷), by magnetically labeling the blood as the blood enters the brain, a technique known as continuous arterial spin labeling.⁹⁸ Patients with large perfusion defects^96,99 detected by PWI or occluded arteries detected by MRA¹⁰⁰ are at heightened risk for enlarging regions of frank infarction, leading some to suggest that these findings should prompt early revascularization, either with thrombolytic agents pharmacologically or with mechanical devices. The volume of abnormalities on DWI and PWI during acute stroke correlates with acute National Institutes of Health Stroke Scales and with chronic neurologic scores, and lesion size may predict early neurologic deterioration.^101-102 Still another benefit of advanced techniques like DWI and PWI is, as is the case for CTPSs, the potential to identify areas of ischemia that have not yet progressed to infarction, potentially permitting extension of the traditional 3-hour window for thrombolytic administration.^103,104 However, PWI and DWI have yet to prove practical and reliable in defining the ischemic penumbra and infarct core,¹⁰⁵ and head-to-head trials comparing MRI diffusion–perfusion studies to CTPSs are limited.

Figure 1-42 shows images from a single patient in various modalities, including CT, CTA, MRI, and MRA.

Ultrasonography for Ischemic Stroke

Ultrasound techniques include Doppler (used to assess flow rate and presence of stenosis), brightness mode (permitting anatomic and structural details of the tissue to be illuminated), and duplex (a combination of the two). Carotid duplex ultrasound has traditionally been deployed electively (i.e., nonemergently) to investigate whether the origin of an acute ischemic event in a given patient could be due to carotid artery stenosis. Studies show conflicting results, with some showing poorer performance of ultrasound (65% sensitivity, 95% specificity) compared with MRA (82%-100% sensitivity, 95%-100% specificity) and the gold standard (by definition 100% sensitive and specific) of digital subtraction angiography.^106,107 Transcranial ultrasound can be used to visualize the vessels in and near the circle of Willis. Here, it is possible to identify stenosis with reasonable success, though less well for the vertebral–basilar system. In the internal carotid artery distribution, the sensitivity and specificity for stenosis are 85% and 95%, respectively. Sensitivity and specificity are only 75% and 85%, respectively, in the vertebral–basilar system. Additional benefits of transcranial ultrasound reportedly include the ability to identify collateral pathways, visualize in real time harmful emboli and in the postthrombolysis state, judge the success of therapy.^108-110 In addition, ultrasound is being used therapeutically in trials to augment the thrombolytic effect of medications.¹¹¹

Studies have focused on the use of ultrasound to select patients for thrombolytic drug therapy or endovascular treatment. Ultrasound is inexpensive relative to other imaging modalities, noninvasive, and does not expose the patient to ionizing radiation, but the validity of the results is highly operator-dependent. In addition, it is impossible to differentiate reliably between complete and high-grade stenosis, and contralateral stenosis can result in falsely reassuring flow velocities ipsilaterally, resulting in a false-negative interpretation for stenosis.^112-115

Conventional Catheter Angiography for Stroke

Still considered the gold standard for diagnosing arterial stenosis, conventional catheter angiography has been substantially improved with the advent of digital subtraction techniques permitting visualization of even small, cortical branches of intracranial arteries. Endovascular techniques permit administration of intra-arterial thrombolysis, and some users have the ability to perform clot retrieval and angioplasty with or without stenting.¹¹⁶ However, despite its utility when noninvasive diagnostic techniques are equivocal or conflicting, angiography is still used only sparingly in the acute stroke setting because of its invasive nature and the approximate 1% risk of iatrogenic stroke associated with the procedure.¹¹⁷

Stroke Neuroimaging Summary

The goals of neuroimaging in the ED patient presenting with signs and symptoms consistent with stroke include the exclusion of intraparenchymal hemorrhage, space-occupying lesions, and other stroke “mimics.” Ideally, the imaging technique would also highlight areas of ischemia, identify underlying causes of ischemia (i.e., vessel occlusion), and delineate those areas that are ischemic but salvageable. Currently, no single imaging modality can accomplish all of these goals quickly and at low cost, and no combination of studies is widely available in all centers.

When stroke is the suspected diagnosis, the patient must be imaged as quickly as possible with the most readily available neuroimaging modality to rule out hemorrhage and other stroke mimics. Typically, this is with noncontrast CT. In the absence of contraindication to the use of contrast, simultaneous CTA can evaluate for large vessel occlusion. Provided that it does not delay other indicated therapy, CTPS may be useful for prognosis and to guide therapeutic decisions, including the use of thrombolytic drug therapy. In specialized centers with the required 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 Is the Risk of Progression to Stroke in Acute TIA? What Imaging Studies Are Indicated?

Large studies in multiple settings have demonstrated that patients diagnosed with TIA in the ED progress to stroke at a high rate of 10% in 3 months, with half of those strokes occurring within 48 hours.^118,120 In addition, some evidence suggests that patients with TIA may be at higher risk for subsequent adverse clinical outcomes than patients with minor ischemic strokes—possibly because of a higher rate of large vessel atherosclerosis as the cause of TIA. In TIA, large artery atherosclerosis (including carotid artery disease) may account for up to 34% of cases. The 3-month risk for stroke, myocardial infarction, and vascular death is higher for TIA patients 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 with TIA or to perform urgent diagnostic testing in the outpatient setting. A variety of neuroimaging strategies have been proposed to identify patients at high risk.

Is Noncontrast CT Valuable in Apparent TIA?

In patients presenting with apparent TIA, head CT might be expected to be normal, and the value of CT in the context of resolved neurologic symptoms could be questioned. Yet 4% of patients clinically diagnosed with TIA in the ED have evidence of new infarct on noncontrast head CT. Moreover, those patients have a fourfold increased 90-day risk of subsequent stroke.¹¹⁸ Thus, head CT may be warranted in TIA as a risk stratification tool. If MRI with or without MRA will be performed in the ED, CT is not necessary as MR can evaluate for hemorrhage, infarct, and other brain pathology.

Is MRI Useful in Apparent TIA?

MRI may be a useful alternative imaging modality when available for patients with apparent TIA. The age of ischemic lesions can be judged with DWI based on measurement of the apparent diffusion coefficient, which varies with time from onset of ischemia. Ischemic lesions of varying ages on DWI MRI predict a substantially higher risk of new lesions on 30-day follow-up MRI (relative risk = 3.6; 95% CI = 1.9-6.8), so MRI may be useful for risk stratification.¹²² Ischemic lesions of varying ages suggest an embolic source of TIA and future stroke. A negative diffusion weighted MRI within 24 hours of apparent TIA predicts a low risk of disabling ischemic stroke within 90 days in patients with moderate or high risk ABCD(2) scores (described later).122a

What Percentage of Patients with TIA or Stroke Have a Cardioembolic Source? Is Echocardiogram Indicated?

Studies show that 51% to 61% of patients with TIA or stroke in whom a clear cause had not yet been identified after initial workup had a cardiac abnormality potentially warranting anticoagulation identified on echocardiography. Transesophageal echocardiography was superior to transthoracic echocardiography, identifying an abnormality in 40% of those with a normal transthoracic echocardiography. An absolute indication for anticoagulation was identified by echocardiography in 20% of patients, and 80% of those were identified only on transesophageal echocardiography.^123,124 Another study demonstrated that coexisting cardiac and carotid artery lesions are quite common in TIA and stroke patients, occurring in 11%, so imaging of both the heart and the cervical vessels is likely indicated.¹²⁵

Does a Clinical Prediction Rule Exist to Identify High-Risk TIA Patients Who Require Further Imaging?

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 disturbance, and duration to generate a score, which has been correlated with 7-day risk for stroke. A dichotomized version of the ABCD score, with patients scoring five or more points being at high risk, has been validated for the ED. Unfortunately, the small numbers in these studies result in wide CIs for the sensitivity of the score, with lower 95% CIs as low as 60%. Some studies have shown the ABCD score to have only limited value, as patients with relatively low risk scores (predicted to correlate with low risk for stroke) had adverse outcomes

Box 1-13 ABCD(2) Score for Predicting Progression from Transient Ischemic Attacks to Stroke^∗

From Asimos AW, Johnson AM, Rosamond WD, et al. A multicenter evaluation of the ABCD2 score’s accuracy for predicting early ischemic stroke in admitted patients with transient ischemic attack. Ann Emerg Med 55:201-210 e5, 2010; Bray JE, Coughlan K, Bladin C. Can the ABCD score be dichotomised to identify high-risk patients with transient ischaemic attack in the emergency department? Emerg Med J 24:92-95, 2007; Cucchiara BL, Messe SR, Taylor RA, et al. Is the ABCD score useful for risk stratification of patients with acute transient ischemic attack? Stroke 37:1710-1714, 2006; Johnston SC, Rothwell PM, Nguyen-Huynh MN, et al. Validation and refinement of scores to predict very early stroke risk after transient ischaemic attack. Lancet 369:283-292, 2007; Rothwell PM, Giles MF, Flossmann E, et al. A simple score (ABCD) to identify individuals at high early risk of stroke after transient ischaemic attack. Lancet 366:29-36, 2005; Tsivgoulis G, Spengos K, Manta P, et al. Validation of the ABCD score in identifying individuals at high early risk of stroke after a transient ischemic attack: A hospital-based case series study. Stroke 37:2892-2897, 2006.

• Age ≥60 years (1 point)

• Systolic blood pressure ≥140 mm Hg and/or diastolic blood pressure ≥90 mm Hg (1 point)

• Unilateral weakness (2 points)

• Speech disturbance without weakness (1 point)

• Duration of symptoms

≥60 minutes (2 points)

10–59 minutes (1 point)

<10 minutes (0 points)

• Diabetes (1 point)

^∗ In validation studies, ABCD(2) poorly predicts short-term stroke risk.

in external validation.^126-129 A recent extended version of the ABCD score including diabetes as a risk factor, known as ABCD(2), attempts to predict 2-day risk for stroke but requires further validation (Box 1-13).¹³⁰ A multicenter study demonstrated a low 7-day risk for progression from TIA to disabling stroke in patients with an ABCD(2) score of no more than three. However, risk for more minor stroke was poorly predicted, limiting applicability in the ED.¹³¹ Another prospective study suggested a 5-fold higher 90-day risk of stroke in patients with an ABCD(2) score >2.131a

What Is the Yield of CTA, Ultrasound, or MRA in TIA?

As stated earlier in the discussion of stroke, CTA has been shown to be 100% sensitive for high-grade carotid stenosis. MRA and carotid ultrasonography are alternative studies with similar sensitivity.¹³²

CT Angiography for Aneurysmal Subarachnoid Hemorrhage

When noncontrast head CT is negative in a patient suspected of SAH, is additional imaging warranted to assess for aneurysmal disease? Is there a role for CTA (see Figure 1-40)? The precise role is as yet undefined. A small study found aneurysms in 5.1% (6/116) of patients with a negative CT and positive LP and in 2.5% (3/116) following a normal CT and LP. Given the incidence of berry aneurysms in the general population, believed to be approximately 1% to 5%,^142,144 it is possible that the aneurysms detected in these patients were incidental, not the acute cause of the patients’ symptoms. Wide use of CTA might result in detection of large numbers of asymptomatic aneurysms, resulting in unneeded procedures, including formal angiography and endovascular coiling of aneurysms, with associated morbidity and mortality. Perhaps the best role of CTA would be in patients in whom LP is not feasible—for example, those with coagulopathy. CTA might also be useful in patients with a particularly high pretest probability of disease but negative noncontrast CT and LP.¹⁴⁵ Finally, CTA could be used to evaluate for aneurysm in patients with negative noncontrast CT and equivocal LP results, such as a declining but nonzero number of red blood cells.

Assessment of Intracranial Pressure Before Lumbar Puncture: Does CT predict elevated ICP before LP?

Some studies have shown size of ventricles on CT to be only weakly predictive of ICP.⁵⁸ Oliver et al. have argued that the evidence that imaging accurately predicts elevated ICP is scant, that some degree of ICP elevation in meningitis is probably ubiquitous, and that examination findings suggesting high ICP, such as stupor, coma, or focal neurologic deficits, are more reliable than CT in identifying patients at risk of herniation.¹⁶⁰ Nonetheless, CT is frequently performed before LP for suspected meningitis, with the goal of ruling out alternative diagnoses and identifying contraindications to LP. A prospective study in 2001 demonstrated an association among age, recent seizure, abnormal neurologic examination findings, and immunocompromise and CT abnormalities before LP, although the majority of the CT abnormalities were felt to be unlikely to contraindicate LP (Table 1-6).¹⁶¹ Because elevations of ICP may be present that are not detected on head CT, LP should be carefully considered in patients with abnormal mental status or neurologic examinations, even if CT appears normal. Conversely, significant midline shift or findings suggesting herniation on CT may rarely be present in patients with normal neurologic examinations. Following head trauma, 1.9% of patients with CT-diagnosed frank herniation and 4.4% of patients with significant brain shift but no herniation had no neurologic deficit in National Emergency X-radiography Utilization Study (NEXUS) II.¹⁶²

TABLE 1-6 Predictors of Head CT Abnormalities Potentially Contraindicating Lumbar Puncture in Patients With Suspected Meningitis

∗ Human immunodeficiency virus or AIDS, immunosuppressive therapy, or transplant.

† Mass lesion, stroke, or focal infection.

‡ Aphasia, dysarthria, or extinction.

Febrile and Typical Recurrent Seizures

Neither emergent nor urgent imaging is recommended for patients with a typical simple febrile seizure (Box 1-15) or recurrent seizures with typical features compared with the patient’s prior seizure history.¹⁶⁸ Do complex febrile seizures require emergency imaging? A recent study found that although 16% of children presenting with new-onset complex febrile seizure had abnormal CT or MR findings, none required emergent intervention (0%; 95% CI = 0%–4%).¹⁶⁹

Do partial seizures suggest a focal structural brain abnormality and thus mandate imaging? A study from South Africa in a region with a high incidence of tuberculosis and neurocysticercosis suggests that the answer is no. This prospective cohort study of 118 children with new-onset partial seizure found abnormal CT scans in only 8 (7%), all of whom were suspected prospectively of having the CT diagnosis. The investigators concluded that routine scanning would require 11 scans and 5 courses of antihelmintic therapy to prevent one case of childhood seizure disorder, versus no scans and 11 courses of drug therapy if all seizure patients were empirically treated.¹⁷⁰ Of course, this study is from a population quite different from the U.S. population, and its results must be viewed in this context.

Head and Brain Imaging in Syncope

Head CT is commonly obtained as part of the evaluation of a patient with syncope, despite little evidence supporting its use. Syncope is a brief, nontraumatic loss of consciousness with loss of postural tone. Syncope is due to global cerebral hypoperfusion, but the brain is generally the “victim” of syncope, not the cause. Syncope is rarely due to stroke, as loss of consciousness requires loss of blood flow to both cerebral hemispheres or to the medullary reticular activating system in the brainstem. Studies of the diagnostic yield of head CT performed for syncope show little pathology clearly related to the syncope episode.¹⁷¹ In one retrospective study, 34%

Box 1-15 Simple Versus Complex Febrile Seizure

From Greenberg MK, Barsan WG, Starkman S: Neuroimaging in the emergency patient presenting with seizure. Neurology 47:26-32, 1996.

• Simple febrile seizure (imaging not generally indicated)

age 3 months to 5 years

generalized

duration <15 minutes

do not recur within 24 hours

• Complex febrile seizure (imaging may be indicated, though recent studies suggest low risk of emergent intervention)

age <3 months or >5 years

focal, with or without secondary generalization

duration >15 minutes

recur within 24 hours

of patients presenting to a community ED underwent head CT, with only 1 patient (0.7%) having the etiology identified by imaging (posterior circulation infarct).¹⁷² A second retrospective study found that 283 of 649 (44%) patients admitted with syncope at two community teaching hospitals from 1994 to 1998 underwent head CT, with 5 (2%) revealing an apparent causal diagnosis: 10 patients underwent MRI without a diagnosis of a cause of syncope. Utilization of head CT fell from 61% to 33% of syncope patients from 1994 to 1998, but diagnostic yield remained extremely low, under 1% for both groups.¹⁷³ A prospective observational study found a 39% rate of head CT in ED syncope patients at Harvard University’s Beth Israel Deaconess Medical Center, with only 5% having head CT abnormalities. That research group proposes that a decision rule for head CT in syncope might reduce utilization by 25% to 50%, although such a rule has not been prospectively validated (Box 1-16).¹⁷⁴ Head CT may be warranted when the history and exam do not fully exclude other diagnoses, such as seizure or stroke, or when trauma results from a syncope episode.¹⁷⁵ A 2007 ACEP Clinical Policy found that there is no evidence for routine screening of syncope patients with advanced imaging including CT. That policy gives a Level C recommendation for cranial CT only when indicated by specific findings in the history or physical examination.^175a

Posterior Reversible Encephalopathy Syndrome (PRES)

Posterior Reversible Encephalopathy Syndrome (PRES) is a state of vasogenic brain edema seen in severe hypertension, eclampsia, lupus, and cyclosporine toxicity, among many other conditions. The mechanism remains

Box 1-16 Proposed Decision Rule for Computed Tomography Head in Syncope^∗

From Grossman SA, Fischer C, Bar JL, et al: The yield of head CT in syncope: A pilot study. Intern Emerg Med 2:46-49, 2007.

• CT head indicated only for patients with one or more of the following

Signs or symptoms of neurologic disease, including headache

Trauma above the clavicles

Coumadin use

Age >60 years

^∗ Not yet validated.

in debate, and studies of imaging findings are limited by the lack of a strong diagnostic reference standard. CT and MR can demonstrate focal regions of symmetric hemispheric edema. The most commonly involved regions are the parietal and occipital lobes, followed by 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 be involved. Complications such as hydrocephalus and hemorrhage can occur. On MRI, restricted diffusion representing infarction is seen in 11% to 26% of cases. The clinical importance of imaging for the specific diagnosis and management of PRES is not well established, though imaging to rule out hemorrhage, mass or mass effect, hydrocephalus, and ischemic stroke is indicated.175b

Epidemiology of Traumatic Brain Injury

The incidence of an acute intracranial injury seen on CT following a “mild” traumatic brain injury (GCS score = 13–15) is approximately 6% to 9%, but not all detected injuries result in a clinically meaningful change in management.^63,176 Of patients in the derivation phase of the Canadian CT Head Rule (CCHR), 8% had potentially important CT head findings, yet only 1% underwent a neurosurgical intervention.⁶³ NEXUS II enrolled 13,728 patients at 21 medical centers in the United States, including all ED patients undergoing head CT after

blunt head trauma, regardless of GCS score or neurologic examination. Of these, 6.7% to 8.7% had clinically significant traumatic blunt head injury (prospectively defined) (Box 1-17).^177,178 For purposes of NEXUS II, a clinically significant traumatic brain injury was defined based on prior research as an injury that may require neurosurgical intervention (such as craniotomy, invasive ICP monitoring, or mechanical ventilation) or an injury with potential for rapid deterioration or long-term neurologic dysfunction.¹⁷⁹ Based on NEXUS II, an average emergency physician evaluating patients with a range of GCS scores and neurologic examination findings might expect to find a potentially important head CT abnormality in between 1 in 20 and 1 in 10 patients in whom CT head was ordered after blunt trauma.^177-178 Stiell and collaborators in Canada showed a 6% incidence of head injuries in ED patients undergoing CT following blunt head injury, although they also demonstrated great heterogeneity in the ordering practices of ED physicians, from 6.5% to 80% of patients with head trauma undergoing CT, depending on treating physician.¹⁸⁰ As described later, a variety of decision rules have been investigated to reduce unnecessary imaging in patients with a very low risk for intracranial injury.

Skull Films for Blunt Head Trauma

The 2008 ACEP clinical policy on altered mental status and mild blunt head trauma found that “plain films of the skull have essentially no utility” in the emergency evaluation of patients with altered mental status.¹⁸¹ Surprisingly, they are still widely used in international emergency practice. In the United Kingdom, skull radiographs are obtained in about 20% of blunt head injury patients, a decrease from nearly 50% in the past.¹⁸² Some investigators have argued that this diagnostic test may play a limited role when CT is not available.^183-184

Clinical Decision Rules for Blunt Head Trauma

Clinical decision rules have been derived by several groups in the United States, Canada, the United Kingdom, and Scandinavia with moderate success. These rules differ in their definitions of clinically significant injuries, the neurologic inclusion criteria (GCS score, loss of consciousness, and neurologic examination), and the time from injury to imaging. Table 1-7 compares three rules, while Boxes 1-19 through 1-23 and Table 1-8 list the specific inclusion and exclusion criteria 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 seek to identify injuries most likely to require neurosurgical intervention or to result in serious neurologic deficits. It is a matter of debate as to which definition is most appropriate. For example, is it important to know of the existence of a traumatic brain injury that does not require neurosurgical intervention to follow cognitive function long-term? Moreover, some clinical interventions may not have truly been “needed” but rather may have been driven by the judgment of individual physicians once they became aware of imaging findings.

TABLE 1-7 Outcomes and Test Characteristics of Three Clinical Decision Rules for Computed Tomography Use Following Minor Blunt Head Injury in Adults∗

TABLE 1-8 Canadian Computed Tomography Head Rule: International Survey Responses

Canadian Computed Tomography Head Rule

The CCHR was 100% sensitive (95% CI 92%-100%) and 68.7% specific (95% CI 67%-70%) for patients with blunt trauma and a GCS score of 13-15 in its original study; subsequent studies have found slightly lower values.^63,185-189 This rule (see Box 1-21) has been criticized for its relative complexity, as well as for end points (see Box 1-19) that

Box 1-19 Canadian Head CT Rule: Outcomes

From Stiell IG, Wells GA, Vandemheen K, et al: The Canadian CT Head Rule for patients with minor head injury. Lancet 357:1391-1396, 2001.

Primary Outcomes: Neurosurgical Intervention

• Death within 7 days secondary to head injury

• Any of the following procedures within 7 days:

Craniotomy

Elevation of skull fracture

ICP monitoring

Intubation for head injury (demonstrated on CT)

Secondary Outcomes

• Clinically important brain injury on CT

Any acute brain finding revealed on CT and that would normally require admission to hospital and neurosurgical follow-up

• Not clinically important

Patient was neurologically intact and had one of the following:

Solitary contusion <5 mm in diameter

Localized subarachnoid blood <1 mm thick

Smear SDH <4 mm thick

Closed depressed skull fracture not through the inner table

might be considered unacceptable in some medical practice settings, including the United States, where fears of litigation might make any CT abnormality undesirable to be missed. Nonetheless, in multiple validation studies and subanalyses, the rule appears to perform well in identifying patients who present with normal mental status but require emergent neurosurgical intervention, including in U.S. populations.¹⁹⁰ Remarkably, despite numerous studies in multiple settings, emergency physician awareness of the rule remains low. A recent study found that only 35% of U.S. emergency physicians were aware of the rule (see Table 1-8).¹⁹¹

Is One Rule “Best”?

A retrospective study of 7955 patients with traumatic head injury published in 2009 compared the three rules we’ve discussed in detail, plus the guidelines of the Neurotraumatology Committee of the World Federation of Neurosurgical Societies, the National Institute of Clinical Excellence (UK), and the Scandinavian Neurotrauma Committee. No statistical difference was found when comparing the sensitivity of the rules for detection of intracranial hematoma requiring surgical treatment. 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

Can We “Mix and Match” the Rules?

It may be tempting to adopt a combination of features of the preceding decision rules to manufacture a superior decision rule. Unfortunately, there is little evidence to support this practice. The NEXUS investigators and the New Orleans investigators tested a variety of other combinations of clinical criteria besides the final suggested rules. ^176,177 Eliminating criteria not surprisingly improved specificity but impaired sensitivity, and adding additional criteria improved sensitivity but impaired specificity. A rule with large numbers of criteria fails the original purpose of developing a clinical decision rule; it detects all injuries by mandating CT for all blunt head trauma patients.

Some Commonalities Among the Rules

An important take-home point for all of the decision rules described earlier is the notable absence of loss of consciousness alone as an indication for head CT. In the CCHR and New Orleans Criteria studies loss of consciousness was a required inclusion criterion, but in none of the studies did isolated loss of consciousness identify patients at risk for significant injury. NEXUS II included patients with or without loss of consciousness in the study population, and did not find that loss of consciousness required head CT in the absence of the other rule critera. Prior studies had shown little association between loss of consciousness and significant traumatic brain injury.¹⁹⁷ Neither NEXUS II nor the CCHR found posttraumatic headache to be an indication for head CT. The New Orleans Criteria did identify headache as a high-risk criterion, though a decision rule omitting headache would have had 97% sensitivity in the derivation phase.¹⁷⁶ Because the New Orleans investigators desired 100% sensitivity, they did not further validate such a rule. Single posttraumatic seizure also does not appear to be a high-risk feature by the CCHR or NEXUS II, though again, it is considered high risk by NOC.

Table 1-9 summarizes the rules for ease of comparison and application.

TABLE 1-9 Criteria for Three Clinical Decision Rules for Computed Tomography Use Following Blunt Head Injury in Adults^∗

Decision Rules for Children

Clinical decision rules for children following blunt head injury remain controversial.¹⁹⁸ Predictors of intracranial injury such as scalp hematoma lack specificity, resulting in large numbers of head CT performed to detect an injury. For example, only about 1% of patients under the age of 2 years with scalp hematoma have traumatic brain injury, although large hematomas, nonfrontal location, and accompanying skull fracture increase the risk.¹⁹⁸ In general, acceptably high sensitivity in these studies comes at the price of extremely low specificity, resulting in little reduction in the utilization of CT in pediatric populations. In some settings, application of these rules could actually increase utilization. These rules may also be very sensitive to the care with which they are applied; the NEXUS II rule, for example, falls in sensitivity from approximately 99% to only 90% when the word headache is replaced with severe headache, so emergency physicians must scrupulously apply the rules if they expect the rules to function as described in the original studies.^199-203

Palchak et al.²⁰⁰ prospectively derived a clinical decision rule for pediatric patients using a cohort of 2043 patients at a single center. They excluded patients with trivial trauma such as falls from standing. They identified five predictors, any of which would mandate CT: abnormal mental status, clinical signs of skull fracture (including retroauricular ecchymosis, CSF otorrhea or rhinorrhea, and palpable fracture), history of any vomiting, scalp hematoma (in children ≤2 years of age), or headache. The sensitivity of the rule was 99% (95% CI = 94%-100%) for traumatic brain injury and 100% (95% CI = 97%-100%) for detection of injury requiring acute intervention. This rule would have eliminated 24% of the CT scans performed at the study institution, but the CIs for the result remain wider than would be accepted by some clinicians and parents.

In the largest study of pediatric blunt head injury to date, Kuppermann et al.²⁰⁴ conducted a prospective study of 42,412 children ages 18 years and younger to derive and validate a clinical decision rule identifying children with extremely low risk for clinically important traumatic head injuries after blunt trauma.²⁰⁴ Of these, 14,969 (35.3%) underwent head CT, with clinically important head injuries occurring in 376 (0.9%), and neurosurgery occurring in 60 (0.1%). The investigators identified two rules: one for children younger than 2 years and a second for children 2 years and older. For children under 2 years of age, the rule had a negative predictive value of 100% (95% CI = 99.7%-100%) and sensitivity of 100% (95% CI = 86.3%–100%). In children 2 years and older, the negative predictive value was 99.95% (95% CI = 99.81%-99.99%) and sensitivity was 96.8% (95% CI = 89.0%-99.6%). The rules would have eliminated the need for imaging in 24.1% of children younger than 2 years and 20.1% of older children, without missing any children requiring neurosurgery. Thus children meeting all low-risk criteria do not require a head CT. Conversely, children who do not meet all of the low-risk criteria do not necessarily require an immediate head CT; the rules had positive predictive values of less than 2.5% for clinical important traumatic brain injury. This underscores a major limitation of clinical decision rules: high sensitivity inevitably comes at the price of poor specificity and potential for overuse. While the Kuppermann rules identify children with very low risk of important brain injury, they are not intended to identify children at high risk. The authors consequently recommend CT or clinical observation in those patients not meeting all low-risk criteria.

In the youngest patients, preverbal children 0 to 3 years of age, clinical assessment is particularly difficult. A very low threshold for CT must be applied after any head trauma, especially considering the possibility of nonaccidental trauma.

Decision Rules for Elderly

The elderly have a high rate of injury with few clinical predictors.²⁰⁵ NEXUS II found that the rate of clinically significant injury in those 65 years of age or older was 12.6%, compared with 7.8% in those younger than 65 years.¹⁷⁸ The CCHR classifies patients older than 65 years as high risk and therefore in need of imaging in the case of traumatic loss of consciousness.

Is a Repeated Head CT Required for Patients With Abnormal Head CT After Blunt Trauma?

A range of reported progression in CT findings has been published, with few clear-cut indications for safely omitting repeat scan. A systematic review from 2006 found that the range of reported progression of injury on repeat head CT varied from 8% to 67%, with resulting neurosurgical intervention in 0% to 54% of patients.²⁰⁶ The review’s authors cite a variety of explanations for this dramatic variability in study outcomes, including selection bias, spectrum bias (studies with more severely injured patients being more likely to show unfavorable outcomes), and poor definitions of injury progression. Risk factors including coagulopathy, poor GCS score, and high overall injury severity appear to be associated with worsening CT abnormalities, but methodologic flaws in the studies reviewed make more specific recommendations impossible. Since the

publication of that review, a prospective study of level I trauma center patients with an abnormal head CT has addressed some of the issues raised by that review. The study stratified patients according to GCS (mild: GCS = 13-15; moderate: GCS = 9-12; and severe: GCS <9) and indication for repeat CT (routine vs. indicated by neurologic deterioration). Among patients undergoing CT for neurologic deterioration, a medical or surgical intervention followed CT in 38%. In contrast, among patients undergoing routine repeat CT, 1% underwent an intervention—in both cases, in patients with a GCS score below 9. The authors conclude that repeat CT is warranted in any patient with neurologic deterioration and routine repeat CT may be warranted among patients with a GCS score below 9. No interventions occurred in patient with a GCS score of 9 or higher undergoing routine head CT, but this study is too small to conclude with certainty that routine repeat CT is never necessary in this group.²⁰⁷ Other recent retrospective studies also suggest that routine repeat head CT is not likely to change clinical management in the absence of a deteriorating neurologic examination, but a larger prospective study will be needed to more stringently define those patients with abnormal CT after blunt trauma in whom repeat CT can be deferred.²⁰⁸

What Is the Best Imaging Modality for DAI?

As the name implies, DAI is damage to white matter tracts throughout the brain, thought to occur as the result of shearing from rapid deceleration, often with a rotational component.²⁰⁹ There is some debate as to the clinical scenarios in which this injury occurs, with some arguing that DAI is a feature only of severe injury while others suggesting it as a mechanism underlying postconcussive syndromes in patients with normal CT.²¹⁰ Studies in patients with mild head injury are problematic, as it is unclear whether MR abnormalities are truly evidence of CT-negative DAI or, rather, false-positive MR findings. CT is generally thought to be poor in detecting these changes, though a gold standard for comparison is often lacking or limited to comparison with MRI. A prospective study in 1988 compared CT and MR for identification 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 DAI.²¹¹ Subsequent studies have often compared new MR image sequences such as fluid attenuated inversion recovery and DWI to other MR sequences, using as a gold standard a clinical definition of DAI (loss of consciousness persisting more than 6 hours after injury, no hemorrhage on CT) plus imaging criteria (presence of white matter injury on MRI). This type of study, in which the gold standard or reference used to determine the accuracy of the experimental test incorporates that test, suffers from incorporation bias.²¹² The sensitivity of MRI may be overestimated as a result.