Platelet-activating factor (PAF) is overproduced in ischemic brain. Although postischemic PAF antagonist administration protects the mature brain in some models, little is known about the effects of PAF antagonists in the immature brain. We hypothesized that the PAF antagonist BN 52021 would attenuate perinatal cerebral hypoxic-ischemic injury. To elicit focal hypoxic-ischemic brain injury, 7-d-old (P7) rats (n = 111) underwent right carotid ligation, followed by 2.5-3.25 h of hypoxia (fractional concentration of inspired O2 = 0.08). BN 52021 neuroprotection was evaluated in three groups of experiments: 1) 25 mg/kg/dose, 0 and 2 h posthypoxia; 2), 25 mg/kg/dose immediately before and 1 h after hypoxia; and 3) posthypoxia-ischemia treatment with BN 52021 12.5, 25, or 50 mg/kg/dose in 2 doses 0 and 2 h after hypoxia. All experiments included concurrent vehicle-injected controls. To quantitate severity of injury, bilateral regional cross-sectional areas (groups 1 and 2) or hemisphere weights (group 3) were evaluated on P12. Both pre- and posthypoxic treatment with BN 52021 (25 mg/kg/dose, two serial doses) decreased the incidence of cerebral infarction from 90% to about 30% (p < 0.02, Fisher's exact test). Measurement of cross-sectional areas confirmed neuroprotection and indicated some benefit of pre- over posthypoxic-ischemic treatment in hippocampus and cortex. Over the dose range tested, the neuroprotective effect of BN 52021 administration was not dose-dependent. In contrast, BN 52021 did not attenuate N-methyl-D-aspartate-induced hippocampal excitotoxic injury in P7 rats. Either prophylactic or "rescue" administration of PAF antagonists decreases the incidence and severity of brain injury associated with an episode of perinatal cerebral hypoxia-ischemia.