Non-invasive combined surrogates of remifentanil blood concentrations with relevance to analgesia

Jörn Lötsch, Carsten Skarke, Jutta Darimont, Michael Zimmermann, Lutz Bräutigam, Gerd Geisslinger, Alfred Ultsch, Bruno G. Oertel

Surrogates may provide easy and quick access to information about pharmacological parameters of interest that can be directly measured only with difficulty. Surrogates have been proposed for opioid blood concentrations to replace invasive sampling, serving as a basis for target-controlled infusion systems to optimize analgesia. We aimed at identifying surrogates of remifentanil steady-state blood concentrations with relevance for its clinical, in particular, analgesic, effects. A "single ascending dose" study design assessed concentration-dependent effects of remifentanil in a double-blind randomized fashion in 16 healthy volunteers. Remifentanil was administered by means of computerized infusion aimed at steady-state effect-site concentrations of 0, 1.2, 1.8, 2.4, 3, 3.6, 4.8, and 6 ng/ml (one concentration per subject, two subjects per concentration). Arterial remifentanil blood concentrations were measured during apparent steady state. Pharmacodynamic parameters were measured at baseline and during steady-state conditions. Potential surrogate parameters included the pupil diameter, the amplitude of pupil light reflex, and the performance in a visual tracking task. Clinical parameters were analgesia to experimental pain, nausea, tiredness, and visual acuity. Remifentanil blood concentrations were well predicted by its effects on the pupil light reflex amplitude, better than by its miotic effects. However, the best prediction for both remifentanil blood concentrations and analgesic effects was obtained using a combination of three surrogate parameters (pupil diameter, light reflex amplitude, and tracking performance). This combination of pharmacodynamic parameters provided even better predictions of analgesia than could be obtained using the measured opioid blood concentrations. Developing surrogates only for opioid blood concentrations is insufficient when opioid effects are the final goal. Combining pharmacodynamic surrogate parameters seems to provide a promising approach to obtain acceptable predictions of relevant clinical effects, with better results than obtained with measuring or estimating blood concentrations.