Monitoring Anesthetic Depth Modification, Evaluation and Application of the Correlation Dimension

Abstract

Anesthesia is administered to patients to facilitate surgical and diagnostic procedures. The anesthesiologist generally determines the amount of anesthetics needed on the basis of body weight. However, the inter-individual variation in sensitivity to anesthetics is wide and the needed level of anesthesia is determined by the strength of surgical stimulation. Therefore the administration of anesthetics should be related to the anesthetic effect and a constant readjustment of the 'depth of anesthesia' (DOA) is necessary. Such adjustments of DOA can only be performed accurately if DOA is monitored. In clinical practice the monitoring is based on the evaluation of multiple indirect physiologic parameters. To provide a more direct image of the effects of anesthetics an analysis of the processes in the Central Nervous System (CNS) is needed. The electroencephalogram (EEG) measures the overall activity of the CNS. Different anesthetics prove to have different effects on the EEG. The difficulty of interpretation of the unprocessed EEG makes its use for routine monitoring of DOA clinically unfeasible. Processing of EEG as applied so far (frequency bands, spectral edge frequency etc.) also is not universally applicable. The impression that the EEG informs directly about changes in the physiological state of the patient and the inadequacy of linear measures so far to assess DOA under different circumstances were the main reasons to apply non-linear mathematics to the EEG. Chaos-theory provides new possibilities to analyze the EEG. One variable from chaos-theory that is promising to be representative for the amount of sub-processes or the complexity of the activity of the CNS as reflected in the actual EEG signal: namely correlation dimension (D2) was selected. This thesis describes an adapted algorithm to compute D2 and evaluates its usefulness to the monitoring of DOA.