Clinical Evaluation of Multi-modality Brainstem Evoked Potentials as a Monitor for Severe Head Injury

Both brainstem auditory evoked responses (BAER) and short latency somatosensory evoked responses (SSER) were recorded to detect brainstem damage in 40 cases (group III), and the results were compared with those in the other two groups; one group (group I) consisted of 64 cases in which BAER was recorded at emergency admission and the other group (group II) consisted of 40 cases in which BAER was repeatedly recorded for several days after emergency admission. All of them suffered from severe head injury having a Glasgow Coma Scale of less than 7. As authors previously reported, that even a single recording of BAER was able to estimate the severity of primary or secondary brainstem damage, better than by the use of neurological findings or computed tomography findings such as deformation of the cistern around the brainstem. Its accuracy of predicting the outcome or recovery from the unconscious state was less than 85%, as shown by the results in group I. The accuracy became 92%, if BAER was repeatedly recorded until 2 weeks after injury, as shown by the results observed in group II. In group III, monitoring of both BAER and SSER at emergency admission and subsequent 2 to 4 days estimated brainstem lesion of all levels including the cervical spinal cord more accurately and the abnormalities of both responses predicted poor recovery from unconsious state following severe head injury (accuracy 99%). It is concluded that the analysis of the multi-modality brainstem evoked responses (BAER and SSER) is the most useful method for predicting the outcome from a severe head injury.