Assessment of analgesia by evoked cerebral potential measurements in humans.

This paper gives a brief account of new methods for the evaluation of pain and analgesia in healthy male volunteers by use of evoked cerebral potentials. In pain research, the so-called late components with latencies between 100 and 400 ms are usually used, but these potentials are non-specific, depending on a variety of factors e.g. on the activity of the ongoing electroencephalogram (EEG) on the subject's arousal/attention mechanisms, on the novelty of the stimulus and on its painfulness. Therefore, if evoked cerebral potentials in response to phasic pain stimuli are to be evaluated quantitatively, constant experimental conditions are essential, including subject selection to obtain as uniform a sample of volunteers as possible. Latency variation in evoked potential components in single studies makes signal averaging methods rather inaccurate to predict the effects of weak analgesics upon cerebral potentials in relation to time. Therefore single trial studies have been performed using transformation of post-stimulus EEG activity in terms of frequency. Because of the short duration of the evoked potentials, various parametric spectral estimators have been investigated for their advantages in EEG analysis. Frequency transformation of stimulus-induced cerebral activity by means of the maximum entropy method gives an enormous increase in the power in the 2-4 Hz frequency band with a very constantly located maximum. Examples are given showing that in this way it might be possible to monitor the time course of efficacy of even so-called weak analgesics.