Although low doses of barbiturates are widely believed to increase sensitivity to pain, studies of the electrophysiologic effects of these drugs on the neurons involved in nociception in the spinal cord have detected only depressant effects. The goal of the studies reported here was to quantify the hyperalgesia resulting from low-dose thiopental infusions and to measure the associated concentrations of thiopental in the plasma, brain, and spinal cord. Nociception was measured using the threshold for motor response to pressure stimulation of the tail (nociceptive threshold) and tail flick latency in the rat. Thiopental was administered by intravenous infusions designed to produce plasma concentrations that either slowly increased or remained at a steady state. Plasma and tissue thiopental concentrations were measured by high-performance liquid chromatography. We observed a reduction in nociceptive threshold that was correlated with the plasma thiopental concentration over the range 2-20 micrograms.ml-1 (7.6-76 microM). The relationship was nonlinear. Nociceptive threshold reached a nadir (36% less than control values) at a mean plasma thiopental concentration of 13.7 micrograms.ml-1 (51.9 microM). The steady-state study showed a similar reduction in nociceptive threshold, with an equilibrium plasma thiopental concentration of 7.6 +/- 1.3 micrograms.ml-1 (28.8 +/- 4.9 microM). Concentrations of thiopental in brain and spinal cord samples were 1.7 +/- 0.03 and 3.5 +/- 1.7 micrograms.g-1, respectively. These studies confirm previous reports of hyperalgesia in association with small doses of thiopental. Reductions in nociceptive threshold and tail flick latency were observed in association with spinal cord concentrations of thiopental in a range reported by others to depress the electrophysiologic activity of neurons involved in nociception.