In Reply:-Goto et al. continue an interesting discussion regarding conflicting results among the cited studies. As they reiterate, evidence exists to support the dose dependency of barbiturates on nociceptive behavior. This phenomenon appears to be due to a dose-related progression starting from suppression of descending inhibitory systems (hyperalgesic effect) to suppression of spinal nociceptive mechanisms (analgesic effect). For this reason, differences in behavioral effects of GABAminergic anesthetics that have variable pharmacokinetic/pharmacodynamic (PK/PD) profiles must be interpreted with caution. Goto et al. suggest that, in our study, the lack of analgesic effect with pentobarbital may be due to the possibility that the dose used produced roughly equal suppression of both descending inhibitory and spinal nociceptive systems. Although this is a feasible hypothesis, it should be noted that the purpose of our study paradigm is to measure persistent pain during phase 2, which is thought to be an expression of spinal sensitization produced by noxious stimulation during phase 1. The timing and dosing of drugs used in our study was designed to maximize drug effect during phase 1, in an attempt to prevent spinal sensitization while allowing for anesthetic recovery, so as to eliminate or at least minimize residual drug effect during phase 2. With this in mind, phase 2 pain behavior is interpreted as how effectively the study drug prevented sensitization during phase 1 and less dependent on whether the drug dose was hypnotic or subhypnotic.

Although flinch counting and weighted scores behavioral measurement have both been validated and widely used, Goto et al. emphasize the value of qualitative distinction between different pain behaviors in the rat formalin test. This is an interesting idea, because each of these behaviors may be an expression of diverse subjective sensations and may respond differentially to pharmacologic manipulations.

Given the obvious difficulties in interpreting pharmacologic effects on experimental pain behavior, it is of vital importance that we enhance such behavioral studies with other concomitant measures, such as PK/PD analyses, electrophysiology, molecular biology, and receptor binding assays. Such strategies may provide additional information regarding specific mechanisms of drug action and further aid in clarifying equivocal behavioral findings.

Ian Gilron, M.D., F.R.C.P.(C.), Pain Research Fellow, Departments of Neurology/Neurosurgery and Anaesthesia, McGill University.

Terence J. Coderre, Ph.D., Director, Pain Mechanisms Laboratory, Clinical Research Institute of Montreal, 110 Pine Avenue West, Montreal, Quebec, Canada H2W 1R7.

(Accepted for publication October 31, 1996.)