Preliminary trial of a noninvasive brain acoustic monitor in trauma patients with severe closed head injury

Background: There is no simple way to assess the injured patient after a loss of consciousness. Computed tomographic scanning is required to rule out anatomic injuries, and invasive intracranial pressure monitoring is needed for the patient with severe traumatic brain injury (TBI). We hypothesized that a noninvasive acoustic monitoring system could provide useful clinical data on the severity and progression of TBI. Methods: Twenty-eight consecutive patients with severe TBI and an indication for invasive intracranial pressure monitoring were studied using the Brain Acoustic Monitor (BAM). Monitoring occurred for 1- to 3-hour time periods on the day of enrollment and each day until the patient's condition stabilized. BAM signals were categorized on the basis of amplitude and positive-to-negative deflection ratio, and then compared with the patient's clinical outcome. Results: BAM signal correlated very strongly with clinical outcome: in 27 of 29 sessions with a normal signal, patients were discharged at a Glasgow Coma Scale score > 13, whereas in 36 of 42 sessions with an abnormal signal, the patient either died or left the hospital with a Glasgow Coma Scale score < 9 (p < 0.00001). The correlation between clinical outcome and initial BAM reading was even stronger: 10 of 10 patients with a normal signal did well, as compared with 3 of 18 patients with an abnormal signal. Conclusion: Noninvasive monitoring of the injured brain can discriminate those patients who will have a poor clinical outcome from those who will do well. Further trials of the BAM are indicated.