The possibility that hypothermia during or after resuscitation from asphyxia at birth, or cardiac arrest in adults, might reduce evolving damage has tantalized clinicians for a very long time. It is now known that severe hypoxia-ischemia may not necessarily cause immediate cell death, but can precipitate a complex biochemical cascade leading to the delayed neuronal loss. Clinically and experimentally, the key phases of injury include a latent phase after reperfusion, with initial recovery of cerebral energy metabolism but EEG suppression, followed by a secondary phase characterized by accumulation of cytotoxins, seizures, cytotoxic edema, and failure of cerebral oxidative metabolism starting 6 to 15 h post insult. Although many of the secondary processes can be injurious, they appear to be primarily epiphenomena of the 'execution' phase of cell death. Studies designed around this conceptual framework have shown that moderate cerebral hypothermia initiated as early as possible before the onset of secondary deterioration, and continued for a sufficient duration in relation to the severity of the cerebral injury, has been associated with potent, long-lasting neuroprotection in both adult and perinatal species. Two large controlled trials, one of head cooling with mild hypothermia, and one of moderate whole body cooling have demonstrated that post resuscitation cooling is generally safe in intensive care, and reduces death or disability at 18 months of age after neonatal encephalopathy. These studies, however, show that only a subset of babies seemed to benefit. The challenge for the future is to find ways of improving the effectiveness of treatment.