Background: Movement in response to painful stimulation is the end point classically used to assess the potency of anesthetic agents. In this study, the ability of modeled propofol effect-site concentration to predict movement in volunteers during propofol/nitrous oxide anesthesia was tested, then it was compared with the predictive abilities of the Bispectral Index and 95% spectral edge frequency of the electroencephalogram, pupillary reflex amplitude, and systolic arterial blood pressure. In addition, the relationships between simple end points of loss and recovery of consciousness, and pupillary, hemodynamic, and propofol concentration indicators were studied. Methods: Ten healthy volunteers were anesthetized with an infusion of propofol, which was increased in three equal steps to 21 mg . kg lean body mass(-1). h(-1). After loss of the ability to hold a syringe and of the eyelash reflex, 60% nitrous oxide was introduced and the trachea was intubated without the use of muscle relaxants. The propofol infusion rate then was decreased to 15.4 mg kg lean body mass(-1). h(-1). Ten minutes later, tetanic electrical stimulation was administered to the thigh via needle electrodes: if movement was observed within 1 min, the propofol infusion rate was increased by 1.75 mg . kg lean body mass(-1). h(-1) 5 min after the stimulus; if not, it was similarly decreased. This 15-min sequence was repeated until volunteers ''crossed over'' from movement to no movement (or vice versa) four times. The propofol infusion rate then was increased to 21 mg . kg lean body mass(-1). h(-1), nitrous oxide was discontinued, the trachea was extubated, and the infusion rate was decreased in five equal steps over 50 min. The times at which the eyelash reflex returned and the birth date was recalled were recorded. The electroencephalogram was monitored continuously (FP1, FP2, ref: nasion, ground: mastoid). Measurements of the pupillary response, arterial blood pressure, and heart rate were recorded during induction and awakening, just before and for 5 min after each stimulation. Arterial blood samples were obtained for propofol assay, and propofol effect-site concentrations were calculated at each time. The predictive value of indicators was compared using a new statistic, the prediction probability (P-K). Results: Loss and return of the eyelash reflex occurred at greater propofol effect-site concentrations than either dropping the syringe or recall of the birthday. The propofol effect-site concentration (in the presence of 60% nitrous oxide) predicted to prevent movement after a supramaximal stimulus in 50% of volunteers was 1.80 mu g/ml (95% confidence limits: 1.40-2.34 mu g/ml). The Bispectral Index (P-K = 0.86), 95% spectral edge frequency (P-K = 0.81), pupillary reflex amplitude (P-K = 0.74), and systolic arterial blood pressure (P-K = 0.78) did not differ significantly from modeled propofol effect-site concentration (P-K = 0.76) in their ability to predict movement. Conclusions: Indicators of pharmacodynamic effect, such as the electroencephalogram, pupillary light reflex, and systolic arterial blood pressure, predict movement as well as effectsite concentration during propofol/nitrous oxide anesthesia. Loss and return of the eyelash reflex correspond to a deeper level of anesthesia than syringe-dropping or recall of the birth date.
