Evoked potential findings in mild traumatic brain injury .2. Scoring system and individual discrimination

Objective: To present an objective scoring system for evoked potential (EP) and electroencephalogram (EEG) findings from patients with suspected mild traumatic brain injury (TBI) as described in the preceding report and to rule out potentially confounding variables. Design: Matched two-group comparison design, with a second matched replication group. Setting: Private, ambulatory, nonemergency facility specializing in assessment, pain management, and rehabilitation of spine and related injuries. Patients and Other Participants: Patients were 54 consecutive referrals for computerized EP/EEG evaluation with histories of primarily motor vehicle accidents and symptoms consistent with mild TBI. Normal controls were 27 screened volunteers. Patients were arbitrarily divided into two groups and matched to the controls on the basis of age and gender. Main Outcome Measures: Outcome measures not included in this phase of investigation. A 0- to 10-point, objective scoring system based on EP and EEG differences was developed to quantify degree of electrophysiological abnormality (as defined by similarity to mild TBI ''profile'') in each subject's recordings. Results: Using a cut-off score of 5.0, 100% of normals were correctly classified, while 74% of the initial group of patients and 52% of a replication group were outside the normal range. The patient scores were not attributable to medication effects or poor performance on the P300 condition tasks. Scores were not related to loss of consciousness (LOG) at the time of the accident or to length of time postinjury. Conclusions: A reliable, objective scoring system was developed based on EP and EEG differences found in mild TBI. The scoring system was effective in discriminating individual patient and control recordings with clinically relevant accuracy. The results were not related to medication effects or poor performance on the attention tasks. Together, these two reports present evidence of electrophysiological dysfunction in mild TBI, with sufficient specificity for clinical application.