Opioids, when administered in large doses, produce brain damage, primarily in the limbic system and association areas in rats. This investigation examined the relationship between opioid dose and severity and frequency of brain damage in rats. Forty male Sprague-Dawley rats were anesthetized with halothane/N2O and underwent tracheal intubation, mechanical ventilation, arterial/venous cannulation, and insertion of a rectal temperature probe and biparietal electroencephalogram electrodes. After surgery, halothane was discontinued and O2/N2O 30%/70% was administered for 1 h. Rats were then randomly assigned to one of eight groups. The control group received a loading dose (LD) of 4 mL/kg of 0.9% normal saline solution (NSS) and a maintenance dose (MD) of 4 mL.kg-1.h-1 NSS. The other groups were given fentanyl lypophilized and reconstituted in NSS with the LD ranging from 50 to 3200 micrograms/kg and the MD from 2 to 128 micrograms.kg-1.min-1. After 2 h of fentanyl or NSS infusion; all rats received 100% O2 and, when alert, their tracheas were extubated; after 7 days the rats underwent cerebral perfusion fixation, followed by light microscopic evaluation. Histopathologic lesions (primarily eosinophilic neuron degeneration) were subjectively graded by a pathologist unaware of the experimental treatment; the grades were based on the percentage of dead neurons. There were no lesions observed in the brain areas in any of the control or 200-8 (LD, microgram/kg; MD, microgram.kg-1.min-1) groups. Eleven of 20 rats in the 400-16, 800-32, 1600-64, and 3200-18 groups showed evidence of brain damage primarily in limbic system structures and association areas (P < 0.05). Our data confirm that fentanyl produces limbic system brain damage in rats, and that the damage occurs over a broad range of doses.