Reactive oxygen species-induced reperfusion injury of the brain is an important cause of neonatal morbidity and mortality following perinatal hypoxia-ischemia. Deferoxamine, allopurinol and oxypurinol have previously been shown to be neuroprotective in vivo during or directly after hypoxia-ischemia. To further characterize and more precisely elucidate whether the neuroprotective properties of these agents are mediated via neuronal and glial cells, or whether endothelial cells contribute to this effect, we tested their ability to protect CA1 neurons in organotypic hippocampal slices. Hippocampal slices obtained from 8-day-old rats were cultured for 7 days and exposed to oxygen/glucose deprivation for 50 min, or used as control slices. Cell damage was assessed at 48 h after oxygen/glucose deprivation using propidium iodide staining. At different time points following oxygen/glucose deprivation we administered dizocilpine, 6-cyano-7-nitroquinoxaline-2,3-dione, and alpha-phenyl-N-tert-butyl nitrone for validation purposes. Deferoxamine, allopurinol or oxypurinol were used as test substances. As expected, 89% and 98% protection was demonstrated with dizocilpine present during or during/after oxygen/glucose deprivation resp. alpha-Phenyl-N-tert-butyl nitrone administered during/after oxygen/glucose deprivation provided 44% protection. However, iron chelation with deferoxamine and inhibition of xanthine oxidase by allopurinol or oxypurinol did not confer neuroprotection. The neuroprotective effect of deferoxamine, allopurinol or oxypurinol, as seen in vivo, may be obtained via inhibition of the production of damaging factors by blood born substances or endothelial cells.