The role of middle latency evoked potentials in early prediction of favorable outcomes among patients with severe ischemic brain injuries.

OBJECTIVE: To explore the role of middle latency evoked potentials (EPs) as predictors for favorable outcome in patients with severe ischemic brain injuries by comparing the prognostic ability of short latency somatosensory and auditory evoked potentials (SLSEP and BAEP) with middle latency somatosensory and auditory evoked potentials (MLSEP and MLAEP).
METHODS: MLSEP, MLAEP, SLSEP and BAEP were recorded in 112 patients with severe ischemic brain injuries (Glasgow Coma Scale ≤ 8). Among them, 83 patients suffered from cerebral ischemic stroke and 29 suffered from anoxic-ischemic encephalopathy after cardiopulmonary resuscitation between 1 and 7 days after the onset of stroke. Outcomes were reviewed 6 months later using the Glasgow Outcome Scale (GOS). A GOS score of 4-5 was considered as a good outcome while a score of 1-3 was considered as poor.
RESULTS: By using the prognostic authenticity analysis of predictors for good outcome, at least unilateral N20 of the SLSEP exit and at least unilateral N60 of the MLSEP exit showed the highest sensitivity which was 100% (95% CI: 86.7%-100%). The bilateral normal N60 showed a high specificity of 97.5% (95% CI: 90.4%-99.6%). It also showed the highest positive likelihood ratio of 6.25% (95% CI: 1.28%-30.59%), which was superior to N20 of SLSEP, V of BAEP, and Pa of MLAEP. The analysis demonstrated that the area under the curve for MLSEP grading was the highest (0.838) compared to that of SLSEP grading (0.784), MLAEP grading (0.659) and BAEP grading (0.621).
CONCLUSIONS: Compared with using N20 of SLSEP analysis alone, adding MLSEP improves the outcome prediction in patients with severe ischemic brain injuries. When an outcome is uncertain after initial evaluation using short-latency EPs, MLSEP is valuable to be used from the first week to further improve prognostication in these patients.