The elusive rat model of conditioned placebo analgesia.

Recent research on human placebo analgesia has suggested the need for rodent models to further elucidate the neural substrates of the placebo effect. This series of 3 experiments therefore was performed in an attempt to develop a model of placebo analgesia in rats. In each study, female Sprague-Dawley rats received an L5 spinal nerve ligation to induce a neuropathic pain condition. Each rat then underwent a 4-day conditioning procedure in which an active analgesic drug or its vehicle (unconditioned stimulus) was associated with the following cues (conditioned stimuli): novel testing room (environmental), vanilla scent cue (olfactory), dim incandescent lighting (visual), restraint procedure/injection (tactile), and time of day and injection-test latency (temporal). The analgesics for each experiment were as follows: Experiment 1 used 90 mg/kg gabapentin, experiment 2 used 3mg/kg loperamide hydrochloride, and experiment 3 used 6 mg/kg morphine sulfate. On the following test day, half of the animals received the opposite treatment, resulting in 4 conditioning manipulations: drug/drug, drug/vehicle, vehicle/drug, and vehicle/vehicle. Nociceptive thresholds were assessed with the mechanical paw withdrawal threshold test each day after the conditioning procedure. In all 3 experiments, no significant differences were detected on test day between control and placebo groups, indicating a lack of a conditioned placebo analgesic response. Our results contrast with prior research that implies the existence of a reliable and robust response to placebo treatment. We conclude that placebo analgesia in rats is not particularly robust and that it is difficult to achieve using conventional procedures and proper experimental design.