Toward an Evidence-Based Approach in the Management of Concussion: The Role of Neuroimaging

Considerable debate centers on whether any single symptom or sign may serve as a better indicator of severity of brain injury in cases of concussion. Few prospective radiographic studies on longitudinal volumetric quantitative analysis of brain mass loss appear in the literature, particularly in association with specific posttraumatic symptoms or signs after mild or moderate head injury (1). Noninvasive radiographic tests, such as CT, MR imaging, single photon emission CT, and positron emission tomography, can provide clinically meaningful information regarding both anatomic and biochemical changes that may occur in the brains of patients with postconcussion symptoms. This information, and the results of sensitive neuropsychological tests, may have important applications in the future medical management of concussions. Presently published guidelines regarding the medical management of sports-related concussions (2) and the management of concussions in the emergency department (3) rely heavily on expert opinion and anecdotal case reports. Evidence-based information is needed to validate current concussion management guidelines.

In this issue of the AJNR, Mackenzie et al (page XXX) report the findings of a longitudinal quantitative analysis of brain atrophy in cases of mild and moderate closed head injury. They used an MR-derived measure of brain parenchyma volume to assess differences between a control group and a posttraumatic head injury group based on serial MR images obtained over time. The results of their study suggest a statistically significant decline over time in the percent of brain parenchyma volume in the trauma group compared with that in the control group. Furthermore, brain atrophy was shown to be significantly greater in patients who had loss of consciousness at the time of trauma than in those who did not. The authors also suggest that initial Glasgow Coma Scale scores were not effective predictors of extent of brain atrophy in the mild to moderate closed head injury group. These results have important implications that, if validated in a well-controlled study in a large number of subjects, could alter current definitions of brain injury severity and possibly alter current guidelines and recommendations regarding the management of concussions. In addition, these results may help to explain the unexpected persistence of postconcussion symptoms in a small population of patients with mild traumatic brain injury. Neuroimaging studies, such as that presented by Mackenzie et al, can play an important role in answering the question of whether current measures of brain injury severity have poor sensitivity, specificity, and precision.

Assessment of the severity of brain injury facilitates determination of the prognosis for recovery, as well as the management of the injury. The usual criteria for the assessment of brain injury severity at the time of trauma include the Glasgow Coma Scale score, the duration of posttraumatic amnesia, and the duration of loss of consciousness (4). Neuroimaging findings presently do not play a role in the classification of brain injury severity. Conventionally, brain injuries are classified as mild, moderate, or severe on the basis of these measures. For example, “mild traumatic brain injury” has been defined as head trauma with a Glasgow Coma Scale score of or more than 13, with a posttraumatic amnesia duration of less than 24 hours, and with loss of consciousness, if any, of less than 30 minutes (5). The term mild traumatic brain injury is misleading as a diagnosis, because it includes a spectrum of manifestations that can range from transient mild symptoms to ongoing disabling problems. This definition sets an arbitrary boundary between the classification of a mild traumatic brain injury, presumably with a benign prognosis, and a moderate brain injury. A continuum of progressive brain injury severity exists in terms of pathologic findings and associated clinical signs and symptoms. Pathologic features that may correlate with traumatic brain injury severity that are not included in the conventional classification criteria may include the location and extent of cortical contusions, intracranial hemorrhages, axonal shear injury, and skull fractures. It has been shown that those patients having mild traumatic brain injury with unilateral or multifocal brain lesions shown on CT scans or MR images are more likely to have neuropsychological symptoms after trauma (6, 7). This emphasizes the need to rethink the classification criteria of brain injury severity.

Concussion is a word often used in the medical literature as a synonym for mild traumatic brain injury. Concussion is the most frequent traumatic brain injury treated by clinicians. It has been estimated that 80% of head injuries involve concussion, or mild traumatic brain injury. The physician’s responsibilities in assessing the condition of a patient with a concussion include determining the need for emergent intervention and, in the case of an athlete, offering guidance regarding the ability to safely return to sports play. Concussion may be complicated by cortical contusions, skull fractures, cerebral edema related to the second impact syndrome, intracranial hemorrhage, neuropsychological deficits, and postconcussion syndrome. The risk for complications associated with concussion is increased in those with prolonged loss of consciousness or posttraumatic amnesia or in athletes who prematurely return to sports play. Clinical management guidelines have been developed to assist physicians in the management of concussion. These guidelines, including recent ones published by the American Academy of Neurology (8) and the American Academy of Pediatrics/American Academy of Family Physicians (3), have increased awareness of signs, symptoms, and potential sequelae associated with concussion. However, these guidelines rely heavily on expert opinion and anecdotal case reports. Definitive, evidence-based information is therefore needed to validate current recommendations. MacKenzie et al suggest the importance of loss of consciousness rather than the Glasgow Coma Scale score as a predictor of outcome in cases of mild and moderate closed head injury. Also, they raise the possibility of a trauma-induced apoptosis as a cause of brain atrophy, which may result in the persistence of postconcussion symptoms in a subpopulation of the mild to moderate brain injury group. Ultimately, neuroimaging studies such as those presented by MacKenzie et al, in conjunction with clinical and neuropsychological data, will help to provide the evidence-based information that is needed to clarify current concussion management guidelines and also to clarify the conventional definitions of mild, moderate, and severe traumatic brain injury.

• Copyright © American Society of Neuroradiology