TREATMENT WITH THE COMPETITIVE NMDA ANTAGONIST GPI-3000 DOES NOT IMPROVE OUTCOME AFTER CARDIAC-ARREST IN DOGS

Background and Purpose-We previously showed that treatment with a comp etitive N-methyl-D-aspartate (NMDA) receptor antagonist GPI-3000 (GPI) improved short-term physiological recovery after incomplete global ce rebral ischemia complicated by dense acidosis. We tested the hypothesi s that GPI administered after resuscitation from cardiac arrest would improve a more long-term recovery as measured by neurobehavioral asses sment and neuropathology 4 days after resuscitation. Methods-Anestheti zed dogs were subjected to 7 minutes of cardiac arrest followed by ves t cardiopulmonary resuscitation. Neurobehavioral outcomes were scored daily on a score ranging from 0 (normal) to 500 (worst), On the fourth day, the animals were killed, and neuropathology was evaluated in a b linded manner in the hippocampus and the neocortex by hematoxylin and eosin staining and by determination of percentage of injured neurons, Three groups of animals were treated in a randomized, blinded protocol with either saline (SAL), low-dose GPI (5 mg/kg followed by I mg/kg p er hour for 2 hours), or high-dose GPI (25 mg/kg, followed by 5 mg/kg per hour for 2 hours). Results-The mortality rate was higher in animal s receiving GPI than in saline-treated control animals (4 of 15 deaths in SAL, 6 of 15 in the low-dose GPI group, and 9 of 18 in the high-do se GPI group). Neurobehavioral scores were depressed in GPI-treated an imals compared with saline-treated control animals in a dose-dependent manner, with 96-hour scores of essentially normal (9+/-2) in saline-t reated animals compared with those animals with significant impairment (181+/-47) treated with high-dose GPI. Neuropathological damage in th e neocortex was most severe in GPI-treated animals, with the percentag e of injured neurons dependent on the dose: 8.3%+/-2.7% SAL, 13.2%+/-6 .4% low-dose GPI, and 39.4%+/-10.1%, high-dose GPI, CA1 neuronal damag e was severe regardless of treatment, Conclusions-Contrary to results seen in experimental global and focal cerebral ischemia, in which NMDA receptor antagonism may improve responses to injury, receptor antagon ism with GPI does not improve brain outcome after cardiac arrest and r esuscitation in the dog, Behavioral and histological outcomes both wer e worsened by GPI treatment at two doses, and mortality was higher rel ative to saline control treatment, We speculate that systemic drug eff ects, as well as potential neurotoxicity of the drug under ischemic co nditions, may be responsible for the deleterious outcomes observed in our cardiac arrest model.