Background: One of the problems encountered in assessment of the hypnotic level during anesthesia is the extraction of a consistent and reliable measure online and close to real time. Hemodynamic parameters such as heart rate and blood pressure are not, at least with the traditional single parameter versus time presentation, adequate for ensuring an optimal level of anesthesia, especially when using neuromuscular blocking agents (NMBA). In the literature, it has been demonstrated that auditory evoked potentials (AEP) are able to provide two aspects relevant to determining level of anesthesia: firstly, they have identifiable anatomical significance and, secondly, their characteristics reflect the way the brain perceives a stimulus.
Methods: The aim of this study was to evaluate the AEP index based on a system identification model, the autoregressive model with exogenous input (ARX-model), and to compare it to the classical method, the moving time average (MTA). The ARX enables the extraction within 15-25 sweeps, depending on the signal-to-noise ratio (SNR), whereas MTA typically needs 250-500 sweeps. The hypothesis of the present study was that since the ARX-model extracts the AEP faster than the MTA-model, the former should be able to detect changes during the brief, intense stimulus of endotracheal intubation. Twelve female patients scheduled for gynecological surgery were included in the study. Anesthesia was initiated with thiopentone and maintained with isoflurane and alfentanil. The AEP was mapped into an index (AEP-index) normalized to 100 when the individual was awake and decreasing to an average of 25 during thiopentone induced anaesthesia. The results were compared to those obtained by MTA-extracted AEP.
Results: During tracheal intubation 9 patients showed an increase in the ARX-extracted AEP-index larger than 15, and 6 of these patients showed an increase larger than 25 (mean increase=33, SD=18). The MTA-extracted AEP-index showed only one patient with an increase larger than 15. The ARX-extracted AEP changed significantly faster than the MTA-extracted AEP.
Conclusion: The ARX-extracted AEP-index increases during tracheal intubation. There is a significant difference between the ARX-extracted AEP and the traditional MTA-extracted AEP, in terms of response time. In order to trace short-lasting changes in the hypnotic level by AEP, the AEP should be extracted by a method with a fast response such as the ARX-model.