A review of the effects of reducing brain temperature on ischemic brain injury is presented together with original data describing the systematic evaluation of the effects of brain cooling on brain injury produced by transient focal ischemia. Male spontaneously hypertensive rate were subjected to transient middle cerebral artery occlusion (TMCAO; 80, 120 or 160 min) followed by 24 h of reperfusion. During TMCAO, the exposed skull was bathed with isotonic saline at various temperatures to control skull and deeper brain temperatures. Rectal temperature was always constant at 37 degrees C. Initial studies indicated that skull temperature was decreased significantly (i.e. to 32-33 degrees C) just as a consequence of surgical exposure of the artery. Subsequent studies indicated that maintaining skull temperature at 37 degrees C compared to 32 degrees C significantly (p < 0.05) increased the infarct size following 120 or 160 min TMCAO. In other studies, 80 min TMCAO was held constant, but deeper brain temperature could be varied by regulating skull temperature at different levels. At 36-38 degrees C brain temperature, infarct volumes of 102 +/- 10 to 91 +/- 9 mm3 occurred following TMCAO. However, at a brain temperature of 34 degrees C, a significantly (p < 0.05) reduced infarct volume of 37 +/- 10 mm3 was observed. Absolutely no brain infarction was observed if the brain was cooled to 29 degrees C during TMCAO. Middle cerebral artery exposure and maintaining brain temperature at 37 degrees C without artery occlusion did not produce any cerebral injury. These data indicated the importance of controlling brain temperature in cerebral ischemia and that reducing brain temperature during ischemia produces a brain temperature-related decrease in focal ischemic damage. Brain cooling of 3 degrees C and 8 degrees C can provide dramatic and complete, respectively, neuroprotection from transient focal ischemia. Multiple mechanisms for reduced brain temperature-induced neuroprotection have been identified and include reduced metabolic rate and energy depletion, decreased excitatory transmitter release, reduced alterations in ion flux, and reduced vascular permeability, edema, and blood-brain barrier disruption. Cerebral hypothermia is clearly the most potent therapeutic approach to reducing experimental ischemic brain injury identified to date, and this is emphasized by the present data which demonstrate complete neuroprotection in transient focal stroke. Certainly all available information warrants the evaluation of brain cooling for potential implementation in the treatment of human stroke.