Clinical Studies
Immediate Neurocognitive Effects of Concussion

Michael McCrea, Ph.D., James P. Kelly, M.D., Christopher Randolph, Ph.D., Ron Cisler, Ph.D., Lisa Berger, M.S.W.

OBJECTIVE: To prospectively measure the immediate neurocognitive effects and early course of recovery from concussion and to examine the effects of loss of consciousness (LOC) and posttraumatic amnesia (PTA) on the severity of neurocognitive impairment immediately after concussion.

METHODS: A sports-related concussion research model was used to allow prospective immediate evaluation of concussion. A total of 2385 high school and college football players were studied. Ninety-one players (3.8%) sustained concussions during the study. A brief neurocognitive and neurological screening measure, the Standardized Assessment of Concussion, was used to assess cognitive functioning before the football season, immediately after injury, and 15 minutes, 48 hours, and 90 days after injury.

RESULTS: Standardized Assessment of Concussion scores immediately after concussion were significantly lower than the preseason baseline score and the noninjured population baseline mean, even for injured subjects without LOC or PTA. Subjects with LOC were most severely impaired immediately after injury, whereas those without LOC or PTA were least impaired. Significant impairment was also detected 15 minutes after injury, but all three groups returned to baseline levels of cognitive functioning within 48 hours.

CONCLUSION: These findings are the first to demonstrate not only that a gradient of increasing concussion severity is represented by PTA and LOC but also that measurable neurocognitive abnormalities are evident immediately after injury without PTA or LOC. (Neurosurgery 50:1032–1042, 2002)

Key words: Amnesia, Athletic injuries, Brain concussion, Brain injuries, Unconsciousness

Traumatic brain injury (TBI) is a major public health concern and a leading cause of disability in the United States each year (39). Nearly 90% of the more than 2 million TBIs that occur annually in the United States are classified as mild TBI (MTBI) or concussion (17, 18, 32), and the incidence of MTBI may actually be under-reported, because a large percentage of patients never seek medical evaluation or treatment (16). The cognitive, emotional, and functional effects of MTBI have been extensively studied in recent years. Several reviews (1, 2, 8) of the scientific literature have attempted to clarify the extended natural history of MTBI, as well as to identify injury-related factors (e.g., loss of consciousness [LOC] and posttraumatic amnesia [PTA]) and other variables that may predict outcomes after MTBI (7, 22, 40).

Very few studies have provided objective empirical data on the immediate neurocognitive effects and early course of recovery after MTBI (27, 29, 30, 41). Situational constraints significantly limit opportunities for prospective research and the ability to use standardized methods for assessment of MTBI during the acute phase. The use of standardized mental status and neurological examination methods beyond traditional injury classification criteria is quite uncommon in most trauma settings. Ratings using the Glasgow Coma Scale (38) and other injury-severity grading systems are highly correlated with neuropsychological and psychosocial outcomes after more severe TBI (36) but may not be sensitive to subtle neurocognitive changes that present risks for more severe underlying neurological complications after MTBI (37).

The sensitivity of routine neurological examinations in detecting underlying cognitive abnormalities after MTBI is quite limited and subtle deficits may be overlooked, because there is often little emphasis on detailed neurocognitive assessments (37). In contrast, neuropsychological testing is considered a sensitive thorough method for detecting and characterizing cognitive and behavioral effects after concussion (10). Although an extensive evaluation by a neuropsychologist immediately after injury is not feasible in most acute care settings, neurocognitive testing conducted in the hospital emergency department may allow clinicians to identify patients at risk of persistent symptoms and disability (6).

Investigation of MTBI in organized sports (e.g., football) affords many advantages not typically associated with traditional studies on the immediate effects and natural history of MTBI, including access to a large at-risk population able to undergo preinjury baseline testing, eyewitness accounts of injuries to document LOC and PTA, feasibility of standardized assessment within minutes after injury, and availability of subjects for postinjury follow-up evaluations to document recovery. Several recent studies used formal neuropsychological testing to elucidate the persistent effects of sports MTBI days or weeks after injury, but those studies were not designed to assess the immediate effects of injury and cannot be considered measures of injury severity, because testing was conducted several days after the occurrence of concussion (11, 19, 24–26). A recent review suggested that LOC resulting from sports-related concussion may not be predictive of eventual neuropsychological outcomes (28), but there are few or no empirical data on the neurocognitive effects accompanying LOC and PTA during the acute phase immediately after injury.

The Standardized Assessment of Concussion (SAC) (31) was originally developed to provide clinicians with a more objective, standardized method to immediately assess athletes who sustain MTBIs during competition or practice. Subsequent studies supported the utility of the SAC as a valid reliable method of measuring mental status and neurological abnormalities within minutes after MTBI among athletes (29,30). Using a sports MTBI research model and the SAC, this study represents what we think is the first attempt to prospectively measure the immediate neurocognitive effects of MTBI, to track the immediate and prolonged course of recovery after injury, and to correlate objective test findings with more-established indices of MTBI severity (such as LOC and PTA) for a large sample of injured subjects.

Next section: Patients and Methods