To the Editor:-In an interesting article, Gilron and Coderre [1]demonstrated lack of analgesic effect of systematically administered pentobarbital in the rat formalin test, contradictory to our results that showed that pentobarbital produces preemptive analgesia via activation of GABAAreceptors. [2]To make this issue even more complex, another group of investigators reported hyperalgesic effects of pentobarbital in this test. [3,4]Although the mechanism for this discrepancy is unclear, we raise a few issues missed by Gilron and Coderre.
First, whether pentobarbital produces analgesia appears critically dependent on the dose administered. In our study, the preemptive analgesic state was created by doses of pentobarbital sufficient to produce loss of righting reflex (16–20 mg [centered dot] kg sup -1). [2]However, the smallest dose (10 mg [centered dot] kg sup -1), which did not produce loss of righting reflex in any animal, was slightly hyperalgesic, [2]in agreement with Abbott et al. [3]showing hyperalgesia by a subhypnotic dose (10 mg [centered dot] kg sup -1). In the study by Gilron and Coderre, where pentobarbital failed to show analgesia, even rats receiving the largest dose (20 mg [centered dot] kg sup -1) did not lose righting reflex, indicating that this dose was relatively smaller than our largest dose when the sensitivity of animals to pentobarbital is taken into account. Because pentobarbital is antinociceptive at the spinal level while suppressing the descending inhibitory system supraspinally, [5]we speculate that increasing doses produce a continuum of effects from disinhibition to suppression of the spinal nociceptive system, resulting in hyperalgesia to analgesia.
The second, and perhaps more intriguing, issue is that pentobarbital appears to differentially affect various pain-related behaviors in the formalin test. Injection of formalin provokes several distinct behaviors, including favoring, lifting, licking, and flinching of the injected paw. We used flinching as a measure of pain and demonstrated pentobarbital-induced preemptive analgesia. [2]In contrast, others used the weighted scoring system involving favoring, lifting, and licking, and failed to show an analgesic effect. [1,3,4]In this regard, it is noteworthy that intrathecally administered muscimol, a GABAAagonist, also selectively attenuates formalin-induced flinching while having no effect on the weighted pain score.*
In fact, such differential alterations in formalin-evoked pain behaviors are not unique to agents with GABAAagonist properties. Accumulating evidence suggests that several other classes of agents also cause dissociation among various pain-related behaviors in the formalin test. For example, amphetamine reduces the weighted pain score while leaving flinching unaffected. [3]Nalaxone, although not generally accepted as an analgesic, inhibits licking but does not alter and may increase flinching. [6]It is a matter of great controversy which of these behaviors or their combination is the best measure of formalin-induced pain. [3,6]However, the problem may really lie in the fact that, when dealing with animal models of pain, researchers have predominantly focused on the intensity of pain but have rarely evaluated the quality of pain. Obviously, this approach is too simplistic, because we know pain is a multifactorial phenomenon, at least clinically. That a number of unrelated classes of agents exert such diverse and differential influences on pain behaviors strongly indicate that these drugs may change the quality of pain in a different fashion, resulting in nonuniform alterations in pain behaviors. In our clinical practice, we ask the patient not only “how much does it hurt?” but also “how does it hurt?”. Then, why not ask the same questions to animals? We believe considerations on the quality of pain are essential to thoroughly understand and to add more clinical relevance to animal models of pain. To this end, we believe investigation on the differential effects of pharmacologic interventions on various pain-related behaviors is a highly promising strategy. [7]
Takahisa Goto, M.D., Assistant Professor of Anesthesia; Hidenori Takahashi, M.D., Ph.D., Assistant Professor and Director, Pain Center; Hisashi Yanagida, M.D., Ph.D., Visiting Professor of Anesthesia, Teikyo University, School of Medicine, Ichihara Hospital, 3426–3 Anesaki, Ichihara-shi, CHIBA 299–01 Japan, Electronic mail: RXR04415@niftyserve.or.jp (to TG).
(Accepted for publication, October 31, 1996.)
*Kaneko M, Hammond DL: GABA acting at GABAAreceptors in the spinal cord limits the second, but not the first phase response to formalin in the rat. Presented on August 20, 1996, at the 8th World Congress on Pain, Vancouver, Canada.