A comparison of auditory evoked potentials and spectral EEG in the ability to detect marked sevoflurane concentration alterations and clinical events

Abstract

Background. Level of consciousness monitors can distinguish between consciousness and unconsciousness during anaesthesia induction and awakening. However, this distinction is rarely a clinical problem. What we do need is a peroperative indicator signalling when the anaesthetic depth comes close to awakening, or when it is too deep. We investigated the ability of the Alaris fast extracted AEP (AAI) and the GE Healthcare Spectral Entropy algorithms State- and Response Entropy (SE/RE) to respond to marked changes in sevoflurane concentration during stable surgery and to clinical incidents. Methods. Both monitors were used simultaneously in 9 patients during sevoflurane-based anaesthesia, which at low concentrations was combined with remifentanil. Additionally, most patients had an epidural block. The response of each monitor to sevoflurane concentration alterations within 0.5-1.5 age-adjusted MAC was recorded, mainly during periods with no surgical stimulation, as was the response to stimulation during surgery and at anaesthesia induction and awakening. Off-line, the numbers of correctly detected events were calculated. Results. In total, 114 events were found. The response rate of all events (95% c.i.) was 20-37% and 40-57% for the AAI- and the Entropy-monitors, respectively, P<0.05 (Wilcoxon Matched Pair test). Conclusions. The Spectral EEG monitor performed significantly better, with a larger number of events detected, compared with the AAI-monitor. However, at the best half the number of events was detected. An anaesthetic ceiling effect might to some part explain this finding. Notwithstanding, continuous anaesthetic depth monitoring may add information to low sensitive semi-continuous standard autonomic monitoring.