D. Jordan

Correlation dimension, an EEG complexity measure, reflects increasing and decreasing anaesthetia using data of a multicenter study: 3AP3-9

*Department of Anaesthesiology, Technische Universität München, München, Germany Background and Goal of Study: Correlation Dimension (CD) was introduced to quantify the order/disorder of signals generated by low dimensional chaotic systems [1]. The present investigation was performed to evaluate the capability of the EEG parameter CD in separating consciousness from unconsciousness and the monotonic behaviour of CD in phases of increasing and decreasing anaesthesia. Materials and Methods: EEG data from a study performed in 6 European centres involving 263 adult patients undergoing surgery under general anaesthesia was used. Patients were randomly assigned to one of 11 anaesthetic combinations. The capability of CD in separating consciousness from unconsciousness was evaluated using prediction probability (PK) [2]. For this purpose, parameter values were calculated immediately before and after loss and return of consciousness. The parameter values in phases of anaesthetic increase (until EEG burst suppression) and anaesthetic decrease were analysed computing Spearman correlation coefficients as a measure of monotonic behaviour. A perfect correlation of the parameter and the monotonic decreasing (anaesthetic increase) respectively increasing (anaesthetic decrease) function will result in a correlation coefficient of 1. Results and Discussions: Analysis of PK revealed a value of 0.65 (transition between consciousness and unconsciousness). Spearman correlation coefficients were 0.30 (“deepening” of anaesthesia) and 0.18 (“lightening” of anaesthesia), where positive values denote a monotone relationship of CD to “depth of anaesthesia”. It may be explained by more “order” and less “randomness” in EEG signals with increasing anaesthesia. The results of PK indicate that CD is less adapted to separate consciousness and unconsciousness, because the parameter may be affected by high dimensional EEG signals. Conclusion(s): Based on the challenging data selection involving numerous anaesthetic regimes, CD shows a monotone behavior in phases of anaesthetic increase and decrease but is less useful for separating consciousness from unconsciousness. References: 1 Lai YC et al., Phys Rev E 2002; 65: 1–5. 2 Smith WD et al., Anesthesiology 1996; 84: 38–51. * European Multicenter EEG/AEP Anaesthesia Monitoring Study Group

Sevoflurane-induced changes of field potential activity are detected by Benfords Law: A-385

Conclusion(s): CB1 / mice had significantly lower MAC-values compared to CB1 / mice. Although isoflurane activates GABAA receptors, in CB1 / mice inhibitory GABAergic transmission may be decreased by stress-induced release of ECs and hence, activation of CB1. In contrast, in CB1 / , GABAeffects of isoflurane are not counteracted by ECs and thus more pronounced. The higher heart rate of CB1 deficient mice supports the anesthesia induced activation of the CB1 cannabinoid receptors in the brain stem which has been shown to enhance cardiac vagal tone3. These results suggest that isofluraneinduced anesthesia interacts with the EC system. References: 1 Katona I et al. J Neurosi 2001; 21: 9506–18. 2 Haseneder R et al. Eur J Pharmacol 2002; 451: 43–50. 3 Pfitzer T et al. Br J Pharmacol 2004; 142: 943–52.