[Methods of automatic feedback regulation for anesthesia. Concepts and clinical use].

Dosing of drugs used in general anaesthetic practice is largely based on experience and trial-and-error. From the very beginning, anaesthesiological research has always attempted to supply a rational description of the rules governing the dose-response relationship. During the last few decades it became possible to identify pharmacokinetics as a main constituent of the relationship, since the rate of efficacy is primarily governed by the kinetics of the drug and the dose. On the other hand, the complete nonlinearity of the dose-response relationship can be demonstrated by means of a pharmacodynamic model. Control of anaesthesia by pharmacokinetic-dynamic models may be considered satisfactory as long as there is no pronounced scatter of the model parameters within a given population. Closed-loop feedback control is contrasted to feedforward control that attempts to monitor a given variable within a preset range by feeding back a measured signal to the drug delivery system. The present article reviews the experience and application of automatic feedback control systems used in anaesthesia. It was shown in all cases that adaptive, model-based feedback control is superior to non-adaptive methods. Pharmacokinetic and/or pharmacodynamic models were successfully applied to the servocontrol of volatile anaesthetics, intravenous hypnotics and neuromuscular blocking agents. Over and above these three applications, directly related to anaesthesia, the impact of feedback control on the regulation of blood pressure and blood glucose is reviewed.