John M. Barkley
L. Anne Hayman
Pedro J. Diaz-Marchan
Some authors have described brain injuries based on the mechanism of injury—such as impact injuries versus inertial injuries, or primary versus secondary injuries (3).
The traumatic neuropathology described in this chapter will be divided into intra-axial injuries and extra-axial injuries. Injury of the brain may occur from blunt or penetrating forces. Lesions such as brain contusions or epidural hematomas occur due to direct force, while other injuries, described later, occur due to acceleration/deceleration or rotational forces.
Direct forces may impact on the skull and underlying brain to cause injury to the brain parenchyma. These manifest acutely as brain contusions (Figure 11-2). A contusion may be hemorrhagic (Figure 11-3) or nonhemorrhagic, and all involve the gray matter of the brain. The two main reasons for not giving contrast prior to imaging a patient with acute intracranial injuries are to perform the exam quickly and to look for intracranial blood products.
Acute blood within the brain is bright on a non-contrast CT of the head, as is contrast material. Thus, contrast is withheld when evaluating the brain. The brightness of blood occurs due to the globin moiety of hemoglobin. Dehydration or an increase in hematocrit, may lead to a dense appearance of normal intracranial vasculature. In addition, hemorrhage may also appear denser than expected in these conditions (4).
Everything bright in the brain on CT may not always be due to acute hemorrhage. Calcification is also bright, and presents physiologically in the pineal gland and choroid plexus. In addition, the basal ganglia may exhibit calcification that is usually symmetrical from side to side.
Dystrophic calcifications sometimes occur within the brain parenchyma due to prior infection or trauma. These dystrophic calcifications do not have mass effect or edema associated with them and will not enhance. Hemorrhagic contusions typically have a hypodense (dark) rim around them that represents edema (1). Arterial and venous malformations may also appear bright on CT, but are nontraumatic in nature. These conditions are associated with avid, tubular enhancement. In the case of an arterialvenous malformation, one may detect a nidus of tangled vessels that are associated with a prominent draining vein.
These congenital or acquired vascular malformations are occasionally mistaken for traumatic injuries to the untrained eye. Follow-up imaging such as MRI or angiography are obtained to further characterize these lesions as vascular malformations. Hypodense lesions on CT (dark lesions) usually indicate the presence of ischemia, infarction or contusions that are non hemorrhagic (5).
While bright contusions are hemorrhagic, non-hemorrhagic contusions are hypodense—or darker than the surrounding brain on noncontrast CT images. This is due to local edema and fluid. Small lesions are often nonhemorrhagic or the blood is beyond the resolution of CT and not visible on the image. These types of lesions may be missed on CT. Other dark, nontraumatic lesions such as lipomas, dermoids, or arachnoid cysts can be characterized as such due to the presence of fat or fluid density.
By definition, all brain contusions are intra-axial and involve the brain parenchyma. Specifically, they involve the gray matter of the brain cortex and typically occur in the anterior and basal aspects of the frontal, temporal, and occipital lobes. These regions are most susceptible to contusions because of the bony ridges immediately adjacent to the aforementioned locations. Temporal lobe injuries are adjacent to the petrous bone or greater wing of the sphenoid bone. Frontal lobe contusions are adjacent to the cribriform plate or lesser wing of the sphenoid bone (2).
Contusions occur in 5–10 percent of patients with moderate or severe head trauma (6). They cause local mass effect of surrounding structures and efface adjacent sulci, ventricles and cisterns. When at the brain periphery, they form acute angles with the skull. If the contusion is located immediately adjacent to the site of external trauma, it is a coup injury. If the contusion is opposite to the site of external trauma, it is a contra-coup injury. If the contusions are large enough, they may cause severe mass effect, midline shift or herniation. These large contusions may require neurosurgical intervention to prevent secondary injuries to the remaining brain such as infarction due to vascular compression, hydrocephalus or cranial nerve palsy. Hemorrhages of the deep gray matter nuclei such as the basal ganglia sometimes occurs after severe mechanisms of injury due to rupture of perforating vessels.
In contrast to the CT appearance of hemorrhage — which depends upon the globin moiety of hemoglobin — the MRI appearance of hemorrhage is quite variable. MRI provides much more detailed information, regarding the chronicity of blood products. This is due to the ability of MRI to characterize, in a temporal fashion, the process of blood product degradation. The MRI imaging parameters, the timing of the hemorrhage and the oxidation of iron in hemoglobin all affect the appearance of blood products on MRI.
Acute hemorrhage contains oxygenated hemoglobin, or oxyhemoglobin inside red blood cells. Acute blood within minutes after bleeding is composed of oxyhemoglobin appears similar to fluid (low T1, high T2).
Oxyhemoglobin is isointense (similar) to gray matter on T1-weighting, and hyperintense (bright) relative to gray matter on T2-weighting. Oxyhemoglobin is rarely detected with MRI unless the patient is actively bleeding or is sustaining intermittent bleeds prior to the MRI.
Active bleeding is rarely appreciated on MRI because the exam is obtained in a delayed fashion, after the patient is stable. This is in contrast to CT, where patients are often imaged in the acute setting, and often require emergent therapy while in the gantry. Many of the treatment decisions prior to obtaining an MRI have already been made from the findings seen on noncontrast CT.
Within hours after acute intracranial hemorrhage, blood within the red blood cells becomes deoxygenated(8–72 hr), resulting in conversion from oxyhemoglobin to deoxyhemoglobin. This appears hypointense (dark) on both T1- and T2-weighting. During this phase, hemorrhage may be difficult to detect or qualify based on routine MRI parameters. Gradient echo sequences are often helpful in these instances, because of the magnetic susceptibility artifact caused by the iron moiety in hemoglobin (7).
Magnetic susceptibility occurs due to drastic changes in tissue magnetization at tissue/air or tissue/metal interfaces (1). This entity causes a blooming appearance, or an enlarged hypointense (dark) halo around areas of metal, hemosiderin or air. The iron that becomes deposited within areas of hemorrhage is often detectable months after the injury on gradient echo MRI, when other MRI sequences have normalized (8).
After 3 to 7 days, intracellular deoxyghemoglobin becomes oxidized into intracellular methemoglobin, that appears hyperintense (bright) on T1-weighting and hypointense (dark) on T2-weighting. After 10 to 14 days, the red blood cells are broken down, releasing extracellular methemoglobin—that is, a paramagnetic substance and appears bright on both T1- and T2-weighting.
Other substances that are bright on a T1 weighted image include gadolinium, fat, melanin and proteinaceous substances. Blood products are classically described as one of the causes of hyperintense (bright) foci on a precontrast T1-weighted MRI image. As stated earlier, because an MRI is usually obtained well after the acute insult, blood products are usually in the methemoglobin phase, which is bright on T1.
As white blood cells are mobilized to the site of chronic blood (months to years after hemorrhage), the extracellular methemoglobin is stored as ferritin and hemosiderin. Ferritin and hemosiderin appear hypointense (dark) to gray matter on both T1- and T2-weighting (1, 9–11).
This appears as a dark ring—the hemosiderin ring around areas of old hemorrhage. Gradient echo and susceptibility sequences, as mentioned previously are exsquisitely sensitive to iron or metal. This allows the sequelae of prior hemorrhage to be detected months to years after the inciting injury. In addition, these sequences are good for detectin blood products or hemosiderin within the sulci and subarachnoid spaces—leptomeningeal hemosiderosis. Leptomeningeal hemosiderosis involves deposits of hemosiderin(iron) within the sulci, causing diffuse low signal intensity within sulci on T1 and T2 weighting.
In summary, the appearance of blood products on MRI is quite variable and depends on the timing of imaging in relation to when the injury occurred. The ability of MRI to image the physiologic degradation and evolution of hemorrhage allows one to more accurately describe the age of blood products. To review, hemorrhage evolves from hyperacute oxyhemoglobin (high T2) to acute deoxyhemoglobin (low T1, low T2) to early subacute intracellular methemoglobin (high T1, low T2) to late subacute extracellular methemoglobin (high T1, high T2), and finally to chronic hemosiderin (low T1/T2 ring; magnetic susceptibility).
Original: Brain Injury Medicine. Principles and Practice
Cwinn AA, Grahovac SZ. Emergency CT Scans of the Head: A Practical Atlas. St. Louis: Mosby, 1998; pp. 3–52.
Brant WE, Helms CA. Fundamentals of Diagnostic Radiology, 2nd edition. Baltimore: Williams & Wilkins, 1999, pp. 25–65.
Pearl GS. Traumatic neuropathology. Clin Lab Med. 1998; 18(1): 39–64.
Rauch RA, Bazan C, Larsson EM, Jinkins JR. Hyperdense middle cerebral arteries identified on CT as a false sign of vascular occlusion. AJNR Am J Neuroradiol 1993; 14(3): 669–73.
Furuya Y, Hlatky R, Valadka AB, Diaz P, Robertson CS. Comparison of cerebral blood flow in computed tomographic hypodense areas of the brain in head-injured patients. Neurosurgery.2003; 52(2): 340–45.
Gean AD. Concussion, contusion and hematoma. In: Gean AD. Imaging of Head Trauma. New York: Raven Press, 1994: 75–206.
Johnston KC, Marx WF Jr. Microhemorrhages on gradient echo MRI. Neurology. 2003; 60(3): 518.
Ripoll MA, Siosteen B, Hartman M, Raininko R. MR detectability and appearance of small experimental intracranial hematomas at 1.5 T and 0.5 T. A 6–7-month follow-up study. Acta Radiol 2003; 44(2): 199–205.
Gomori JM, Grossman RI, Goldberg HI, Zimmerman RA, Bilaniuk LT. Intracranial hematomas: imaging by high-field MRI. Radiology 1985; 157: 87–93.
Clark RA, Watanabe AT, Bradley WG, Roberts JD. Acute hematomas: effect of deoxygenation, hematocrit, and fibrin-clot formation and retraction on T2 shortening. Radiology 1990; 175: 201–206.
WeissLeder R, Rieumont MJ, Wittenberg J: Primer of Diagnostic Imaging, 2nd edition. St. Louis: Mosby, 1997, pp. 465–80.