[Brainstem auditory evoked potentials as a method to assist the diagnosis of brain death]

Abstract

Background: brainstem auditory evoked potentials in brain death.

Aim: to verify the agreement between the response in the auditory brainstem audiometry and the clinical outcome, analyzing the pattern of responses to electric stimulation.

Method: a cross-sectional study performed in 30 patients with Glasgow coma score of 3, submitted to the auditory brainstem audiometry and followed up until their clinical outcome: recovery or death. The test was considered positive to brain death when there was no registry of waves or when there was only the registry of wave I; and negative when there were two or more waves, independently of their latencies.

Results: Among the patients who presented positive results for brain death (86.66%), all died; the only patient who recovered presented a negative result, indicating a specificity of 100%. Internal consistency of data was also observed, with an intraclass correlation coefficient of 0.562, obtained using the Cronbach s test; and a significant agreement between the test and the clinical outcome using the Kappa s test, with a confidence interval of 95% (K = 0.545; p = 0.015).

Conclusion: in the present study, the brainstem auditory evoked potential demonstrated to be highly specific in death prediction of patients in Glasgow coma score of 3, and was useful in assisting the diagnosis of brain death.