Altersabhängige Korrelation zwischen EEG-Parametern und zunehmender Narkosetiefe unter Propofol Effekte von Fentanyl

Zusammenfassung

Mit dieser Studie soll der Zusammenhang zwischen klinisch unterschiedlichen Narkosetiefen und drei EEG-Parametern (Median, SEF und ein multivariat berechnetes EEG-Stadium) während einer langsamen Narkoseeinleitung mit Propofol (18 mg/kg/h) bestimmt werden. Es wurden 48 Patienten aus zwei Altersgruppen untersucht. Die Hälfte der Patienten erhielt vor Beginn der Propofolinfusion 0,15 mg Fentanyl. Die klinische Narkosetiefe wurde anhand definierter Zeichen wie z.B. Wegfall des Lidreflexes bestimmt. Insgesamt zeigte sich, daß klinisch unterschiedliche Narkosezustände anhand von EEG-Parametern abzugrenzen sind, wobei das multivariat berechnete EEG-Stadium den Monoparametern Median und SEF teilweise überlegen ist. Dies war unabhängig von alleiniger Propofolgabe oder additiver Fentanylapplikation. Es ergab sich eine deutliche Veränderung des EEG mit zunehmendem Alter, die die Interpretation erschwerte. Fentanyl führte zu einer Verkürzung der Einleitung. Eine Abwehrreaktion auf die Intubation ließ sich nachträglich nur unter bestimmten Bedingungen aus dem EEG ableiten, während das Auftreten von hämodynamischen Reaktionen bei Intubation nicht durch EEG-Parameter abgeschätzt werden konnte.

Abstract

This study was designed to determine the relationship between the electroencephalogram (EEG) and clinical signs of depth of anaesthesia during induction of anaesthesia by slow infusion of propofol (18 mg/kg·h).

Methods. Four groups of 12 patients each were studied (groups I and II: 18–50 years; groups III and IV: >70 years). Groups II and IV were given 0.15 mg fentanyl before the infusion of propofol was started. The clinical signs recorded were: (1) loss of eyelash reflex; (2) respiratory insufficiency; (3) tolerance to painful stimuli; and (4) intubation. Cardiovascular reactions were documented. The dosage was calculated from the infusion time (time from start of infusion until specific clinical event). Bipolar electrodes were placed at the C4/P4 positions (10–20 placement system) to record the EEG, which was processed by a personal computer (Narkograph) using fast-fourier transformation. The Narkograph calculates multiparametric EEG stages ranging from A to F (according to Kugler) as well as median frequency and spectral-edge frequency 95% (SEF). Stage A represents alpha rhythm, stage F is equivalent to a burst suppression pattern. For statistical analysis a Student t-test was performed.

Results. The infusion of propofol led to slowly developing anaesthesia with loss of eyelash reflex followed by loss of pain response, respiratory insufficiency, and intubation. In the younger patients the clinical signs coincided with well-differentiable EEG patterns. Above 70 years of age there were problems in distinguishing the EEG patterns, as there are alterations of the EEG with advanced age. The multiparametric EEG stage calculated by the Narkograph showed a better correlation with the clinical signs than median or SEF. Fentanyl shortened the induction time remarkably: less propofol was needed to achieve corresponding clinical signs when fentanyl was added. The EEG patterns typical for a specific clinical condition remained unchanged by fentanyl. Similar clinical situations showed equal EEG stages in all groups. Different clinical situations could be distinguished by significant changes in the EEG. The infusion times for tolerance to pain and respiratory insufficiency were not significantly different, and there were no significant differences between the EEG patterns and propofol doses for these two clinical parameters. Intubation was performed after 18.5±4.6 min in group I with a propofol dose of 5.6±1.4 mg/kg. This time was shortened by fentanyl in group II to 10.1±3.7 min and a propofol dose of 3.0±1.1 mg/kg.

Conclusion. Different clinical signs corresponding to different levels of depth of anaesthesia could be differentiated by their EEG parameters. The EEG stage allowed better differentiation of the clinical conditions than the single-parameter EEG derivatives median and SEF. The results of this study show that EEG monitoring provides information about depth of anaesthesia.