The imaging of subdural hematoma has evolved significantly. Computed tomography and MRI have supplanted other procedures and rendered most obsolete for the evaluation of intracranial pathology because of ease of use, tremendous soft tissue resolution, safety, and availability. Noncontrast computed tomography has become the accepted standard of care for the initial evaluation of patients with suspected subdural hematoma because of widespread availability, rapid acquisition time, and noninvasive nature. MRI offers important features in determining potential secondary causes of subdural hematoma, such as dural-based neoplasms.
Key points
- •
In the early postoperative period, noncontrast head CT is most commonly used to evaluate for potential complications.
- •
CT is widely available, fast, relatively inexpensive, and accurate at identifying most postoperative complications.
- •
Intracranial air in the early postoperative period can cause artifact on MRI.
- •
CT is sensitive for the identification of new intracranial hemorrhage, new mass effect and herniation, tension pneumocephalus, and calvarial fractures.
- •
Although MRI is less often used in the immediate postoperative period, it is much more sensitive for the detection of acute ischemia and infection.
- •
The choice to use MRI in the postoperative setting should be driven by the clinical scenario; if acute ischemia or infection is suspected clinically, MRI should always be considered.
Background on imaging of subdural hematomas
Historical Imaging Techniques
The imaging of subdural hematoma has evolved significantly. Before modern cross-sectional techniques, such as computed tomography (CT) and MRI, radiography of intracranial pathology generally relied on distortion of normal structures to suggest an intracranial process. Early techniques including plain film radiography, ventriculography, pneumoencephalography, and catheter angiography were limited in the evaluation of the brain parenchyma and surrounding structures given their inherently poor soft tissue resolution. Therefore, the intracranial structures, including the brain parenchyma and ventricles, had the same density on plain film radiography making evaluation for intracranial hemorrhage, tumors, and other intracranial pathologies impossible to visualize. With the exception of plain film radiography, these tests were generally invasive. For example, ventriculography involved drilling a burr hole and directly injecting air into the ventricular system. Pneumoencephalography also relied on similar principles, although the air was instilled into the subarachnoid space of the spine rather than directly into the ventricular system. Both techniques relied on plain film radiography and tomography to evaluate the ventricles to determine ventricular contour irregularities suggesting a mass occupying lesion. These studies were uncomfortable for the patient, often inducing vertigo, nausea, and vomiting. Complication rates were high, and included headache, neck stiffness, meningitis/ventriculitis, altered consciousness, tachycardia, and focal neurologic signs. For these reasons, repeat studies were often not performed given the level of patient discomfort and risks, which limited the evaluation of disease progression over time. Although catheter angiography depicted the cerebral blood vessels with great detail, it too included substantial risks, was plagued by inherently low soft tissue resolution, and also relied on distortion or displacement of the cerebral vasculature to suggest an underlying space-occupying lesion. CT and MRI have supplanted other procedures and rendered most obsolete for the evaluation of intracranial pathology because of ease of use, tremendous soft tissue resolution, safety, and availability.
Current Imaging Recommendations
Noncontrast CT has become the accepted standard of care for the initial evaluation of patient’s with suspected subdural hematoma because of widespread availability, rapid acquisition time, and noninvasive nature. Advanced-generation CT scanners now use multiple detectors, helical acquisition, dual-source, and dual-energy techniques, further improving the quality of CT scans while potentially decreasing the radiation dose to patients depending on the type of study performed. MRI generally has a more limited role in the evaluation of acute intracranial hemorrhage, particularly when evaluating subdural hematoma, for practical reasons including availability in most emergency departments. However, MRI offers important features in determining potential secondary causes of subdural hematomas, such as dural-based neoplasms. There is no longer a role for plain film radiography, ventriculography, or pneumoencephalography in the evaluation of traumatic or nontraumatic intracranial hemorrhage.
Background on imaging of subdural hematomas
Historical Imaging Techniques
The imaging of subdural hematoma has evolved significantly. Before modern cross-sectional techniques, such as computed tomography (CT) and MRI, radiography of intracranial pathology generally relied on distortion of normal structures to suggest an intracranial process. Early techniques including plain film radiography, ventriculography, pneumoencephalography, and catheter angiography were limited in the evaluation of the brain parenchyma and surrounding structures given their inherently poor soft tissue resolution. Therefore, the intracranial structures, including the brain parenchyma and ventricles, had the same density on plain film radiography making evaluation for intracranial hemorrhage, tumors, and other intracranial pathologies impossible to visualize. With the exception of plain film radiography, these tests were generally invasive. For example, ventriculography involved drilling a burr hole and directly injecting air into the ventricular system. Pneumoencephalography also relied on similar principles, although the air was instilled into the subarachnoid space of the spine rather than directly into the ventricular system. Both techniques relied on plain film radiography and tomography to evaluate the ventricles to determine ventricular contour irregularities suggesting a mass occupying lesion. These studies were uncomfortable for the patient, often inducing vertigo, nausea, and vomiting. Complication rates were high, and included headache, neck stiffness, meningitis/ventriculitis, altered consciousness, tachycardia, and focal neurologic signs. For these reasons, repeat studies were often not performed given the level of patient discomfort and risks, which limited the evaluation of disease progression over time. Although catheter angiography depicted the cerebral blood vessels with great detail, it too included substantial risks, was plagued by inherently low soft tissue resolution, and also relied on distortion or displacement of the cerebral vasculature to suggest an underlying space-occupying lesion. CT and MRI have supplanted other procedures and rendered most obsolete for the evaluation of intracranial pathology because of ease of use, tremendous soft tissue resolution, safety, and availability.
Current Imaging Recommendations
Noncontrast CT has become the accepted standard of care for the initial evaluation of patient’s with suspected subdural hematoma because of widespread availability, rapid acquisition time, and noninvasive nature. Advanced-generation CT scanners now use multiple detectors, helical acquisition, dual-source, and dual-energy techniques, further improving the quality of CT scans while potentially decreasing the radiation dose to patients depending on the type of study performed. MRI generally has a more limited role in the evaluation of acute intracranial hemorrhage, particularly when evaluating subdural hematoma, for practical reasons including availability in most emergency departments. However, MRI offers important features in determining potential secondary causes of subdural hematomas, such as dural-based neoplasms. There is no longer a role for plain film radiography, ventriculography, or pneumoencephalography in the evaluation of traumatic or nontraumatic intracranial hemorrhage.
Imaging characteristics
General Features
Subdural hematoma is defined as an extra-axial collection of blood products in the subdural space, which is a potential space between the arachnoid and dura mater. The dura mater is the outermost meningeal layer covering the brain parenchyma. The dura is a thin, fibrous covering that extends over the entire brain and is continuous with the periosteum. The dura is reflected along the medial cerebral hemispheres where two layers form the falx cerebri. Similarly, the dura is reflected along the undersurface of the cerebral hemispheres forming the tentorium cerebelli, which divides the supratentorial and infratentorial compartments. The dural venous sinuses travel between the two dural leaflets along the falx cerebri and tentorium cerebelli. Cortical veins draining the brain parenchyma empty into the dural venous sinuses by crossing the subdural space, hence the term bridging veins. Subdural hematomas are usually caused by tearing of these bridging veins.
Trauma is the most common cause of subdural hematoma. Sudden acceleration/deceleration of the head, rapid head rotation, or direct laceration from skull fractures or penetrating projectiles can tear bridging cortical veins. Spontaneous atraumatic subdural hematomas are particularly common in the elderly and people with alcoholism because of the increased incidence of cerebral atrophy in these populations. Cerebral atrophy causes the extra-axial spaces, including the subdural space, to enlarge and the bridging veins to elongate making them more susceptible to injury. Spontaneous subdural hematomas can also occur in patients with coagulopathy and those taking anticoagulant or antithrombotic medications, such as aspirin, clopidogrel, or warfarin. Less common causes of subdural hematoma include dural-based neoplasms, such as meningiomas, hemangiopericytomas, or metastases. Intracranial hypotension caused by cerebrospinal fluid (CSF) leak following cranial, spinal, or paranasal sinus surgery is another uncommon cause of atraumatic subdural hematoma. Additional rare causes of subdural hematoma include hyponatremic dehydration and dural venous sinus thrombosis. In rare circumstances the subdural hematoma is caused by injury to a cortical artery or ruptured vascular lesion, such as an aneurysm or vascular malformation ( Fig. 1 ). In some circumstances, no clear cause of the subdural hematoma is identified ( Box 1 and 2 ).
- •
Trauma
- •
Spontaneous
- ○
Elderly (with parenchymal atrophy)
- ○
Alcoholics (with parenchymal atrophy)
- ○
Coagulopathic (underlying clotting disorder, liver failure, and so forth)
- ○
Coagulopathic caused by blood thinning medications
- ○
- •
Idiopathic
Dural-based neoplasms (eg, meningioma, hemangiopericytoma, metastasis)
Intracranial hypotension
Hyponatremic dehydration
Vascular lesions (eg, aneurysm, dural arteriovenous fistula, arteriovenous malformation)
Dural venous sinus thrombosis
Anatomic barriers determine the imaging appearance of subdural hematomas. A crescentic extra-axial collection overlying the cerebral convexities is the classic imaging appearance of subdural hematoma. Subdural hematomas often cross calvarial sutures (by contrast with epidural hematomas); however, they rarely cross midline because of continuity of the dural membrane with the falx. Similarly, subdural collections often marginate the falx cerebri medially and tentorium cerebelli inferiorly ( Fig. 2 ). Smaller collections of blood can be seen along either the falx cerebri or tentorium cerebelli alone. Without these anatomic barriers, small extra-axial collections overlying the cerebral hemispheres may be difficult to localize accurately into the correct anatomic extra-axial space: epidural, subdural, or subarachnoid. Subdural hematomas tend to distribute diffusely throughout the potential space between the dura and arachnoid meningeal layers. By contrast, adherence of the dura to the periosteum limits the spread of blood products, giving epidural collections their more focal and biconvex shape. This anatomy also explains why epidural collections do not cross suture lines because the dura mater is densely adherent to the inner table of the calvarium at the sutures. Exceptions to these imaging guidelines may be seen with dural hypoplasia or trauma. Subarachnoid hemorrhage is distinguishable from subdural hematomas by its extension within the subarachnoid spaces, and characteristic presence in cerebral cisterns and sulci. Subdural hematomas can also occur, albeit less frequently, in the posterior fossa and spinal canal. Posterior fossa subdural hematomas have a similar imaging appearance to supratentorial subdural collections ( Fig. 3 ). Subdural hematomas can sometimes be seen in the upper cervical spine on the caudal-most slices of a head CT or MRI. Therefore, careful evaluation of the upper cervical spine at the margin of cross-sectional imaging must be performed to identify spinal lesions.
Computed Tomography Characteristics
The appearance of blood products on CT changes according to their radiodensity as measured in Hounsfield units (HU). This density depends predominantly on the age of the hemorrhage and stage of blood degradation. Therefore, most subdural hematomas follow a fairly predictable course on serial CT scanning. Most acute subdural hematomas are hyperdense compared with the cerebral cortex ( Fig. 4 ). The density decreases by approximately 1.5 HU per day ( Fig. 5 , Table 1 ). Therefore, by Days 7 to 10, the hematoma are approximately isodense to the cerebral cortex ( Fig. 6 ), and by Days 10 to 14 should be hypodense relative to cortex ( Fig. 7 ). Several confounding factors can affect the apparent density of subdural hematomas on CT making age determination inaccurate. These factors include hyperacute blood (active bleeding), coagulopathy preventing clotting, CSF leakage through torn arachnoid membranes that dilutes blood and limits dense clot formation, blood products of varying ages, profound anemia, and superimposed infection ( Box 3 ).
| Substance | CT Density (HU) |
|---|---|
| Air | −1000 to −600 |
| Fat | −100 to −60 |
| Water | 0 |
| CSF | 15 |
| Kidney | 30 |
| Muscle | 10–40 |
| White matter | 20–30 |
| Gray matter | 37–45 |
| Acute blood | 50–80 |
| Liver | 40–60 |
| Bone | >700 |
Unclotted, active bleeding (often in patients with coagulopathy)
CSF leakage
Anemia
Blood products of varying ages
Superimposed infection
MRI Characteristics
Age determination and signal characteristics of subdural hematomas on MRI are more complex and less predictable than CT density. The variable MRI appearance of subdural hematoma is related to hemoglobin structure and its oxidation products. Hemoglobin passes through various forms during red blood cell lysis including intracellular oxyhemoglobin, intracellular deoxyhemoglobin, intracellular methemoglobin, extracellular methemoglobin, and finally extracellular ferritin and hemosiderin. These forms roughly correlate with blood products in the hyperacute (<24 hours), acute (24–72 hours), early subacute (3–7 days), late subacute (7–14 days), and chronic stages (>14 days), respectively ( Fig. 8 ). However, accurate dating of subdural hematoma is sometimes difficult and inaccurate because of heterogeneity of the oxidation progression.
Multiple forms of hemoglobin are often found in a single hematoma. The paramagnetic dipole-dipole interactions of methemoglobin are responsible for the T1 hyperintense signal seen in the subacute phase, whereas magnetic susceptibility effect is responsible for the T2 signal loss (hypointensity) seen in the acute (deoxyhemoglobin), early subacute (intracellular methemoglobin), and chronic (hemosiderin) stages of blood product evolution. Additionally, as the protein content of the hematoma increases over time, so does the T2 hypointensity (see Fig. 8 ). Subdural hematomas are often hyperintense in the acute, subacute, and chronic stages on fluid attenuated inversion recovery (FLAIR) images, whereas the signal characteristics are variable in the hyperacute stage. Hematomas are usually hypointense on gradient echo (GRE), T2*, and susceptibility-weighted images (SWI) caused by magnetic susceptibility effects. GRE, T2*, and susceptibility sequences are exquisitely sensitive for the detection of chronic blood products. The magnetic susceptibility effects are enhanced with higher field strength magnets (3T and 7T), whereas fast spin echo techniques dampen the effects. Subdural hematomas in all stages often contain various amounts of hyperintense signal on diffusion weighted (DWI) images, which is often caused by magnetic susceptibility artifact. DWI hyperintensity should not be confused with superimposed infection, unless clinical circumstances warrants such consideration.
Acute subdural hematoma
Overview of Imaging Characteristics by Time (Days 0–3)
An acute subdural hematoma is defined as being zero to approximately several days old. Acute subdural hematomas are one of the leading causes of morbidity and mortality in patients with severe traumatic brain injury. Traumatic acute subdural hematomas are often accompanied by other types of intracranial hemorrhage, such as subarachnoid hemorrhage, intraventricular hemorrhage, parenchymal contusions, and epidural hematomas.
Noncontrast Head Computed Tomography Findings
Most (60%) acute subdural hematomas are hyperdense relative to gray matter with Hounsfield units typically ranging from 50 to 80. It is helpful to use “wide” (subdural) windowing (window, 175; level, 50) and “narrow” (brain) windowing (window, 80; level, 40) to separate hyperdense blood products from adjacent cortex ( Fig. 9 ). Up to 40% of acute subdural hematomas have a mixed hyperdense and hypodense appearance ( Fig. 10 ). The hypodensity may reflect unclotted blood products, CSF leakage through torn arachnoid membranes, or profound anemia. A “swirl sign,” which appears as irregular pockets of hypodensity within the dominant hyperdense collection, is most often associated with rapid, active bleeding ( Fig. 11 ). In patients with coagulopathy or anemia, the collections may rarely be isodense with respect to underlying brain parenchyma making detection more challenging.
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
