Prognostic value of brainstem auditory evoked potentials (BAEPs) and middle latency auditory evoked potentials (MLAEPs) in unconscious patients after cardiac arrest

Le résumé fourni par la source

AIMS: Assessing neurological outcome after cardiac arrest (CA) remains challenging. We aim to evaluate the prognostic value of brainstem auditory evoked potentials (BAEPs) and middle latency auditory evoked potentials (MLAEP) in patients with disorders of consciousness (DoC) after CA and indeterminate outcome according to current recommendations (i.e., no highly malignant EEG and bilateral absence of N20). METHODS: A retrospective analysis was conducted using data collected prospectively from two intensive care units. The prognostic value of the absence of wave V, an abnormal conduction latency between waves III and V, an abnormal I/V amplitude ratio on BAEPs, and the absence of Na and Pa responses on MLAEPs was assessed. The primary outcome was the best Glasgow Outcome Scale-Extended (GOS-E) level within 3 months. A GOS-E level ≤ 2 (death or vegetative state) was defined as a poor outcome. RESULTS: 93 patients were included, 74.2% of whom were men, with a mean age of 58 years IQR [47-67]. Of these, 67/93 (72%) had a poor outcome. Unilateral absence of wave V on BAEPs was highly specific for poor outcome (Sp 96.1%), despite low sensitivity (Se 10.4%). A unilateral I/V amplitude ratio > 1 had a specificity of 87.5%, but a low sensitivity of 11.5%. Prolonged brainstem conduction time (III-V latency > 2 ms) was poorly specific for a poor outcome (Sp 70.8%). Conversely, the absence of Na and Pa on MLAEP had 100% specificity for poor outcome, with a sensitivity of 31.8%. CONCLUSION: In unconscious patients after CA with an indeterminate outcome based on neurophysiological criteria, the absence of wave V on BAEPs and the absence of Na/Pa on MLAEPs were highly specific for poor outcome, despite a low sensitivity. These results suggest that AEPs may serve as a second prognostic marker to fulfill the two-criterion requirement of the ERC/ESICM recommendations.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.