J Neurosurg 77:337-354, 1992 Review Article Pathophysiology and treatment of focal cerebral ischemia Part 11: Mechanisms of damage and treatment Bo K. SIESJO, M.D. Laboratory for Experimental Brain Research, Experimental Research Center, Lund University Hospital, Lund, Sweden Y The mechanisms that give rise to ischemic brain damage have not been definitively determined, but considerable evidence exists that three major factors are involved: increases in the intercellular cytosolic calcium concentration (Ca++J, acidosis, and production of free radicals. A nonphysiological rise in CaY due to a disturbed pump/leak relationship for calcium is believed to cause cell damage by overactivation of lipases and proteases and possibly also of endonucleases, and by alterations of protein phosphorylation, which secondarily affects protein synthesis and genome expression. The severity of this disturbance depends on the density of ischemia. In complete or near-complete ischemia of the cardiac arrest type, pump activity has ceased and the calcium leak is enhanced by the massive release of excitatory amino acids. As a result, multiple calcium channels are opened. This is probably the scenario in the focus of an ischemic lesion due to middle cerebral artery occlusion. Such ischemic tissues can be salvaged only by recirculation, and any brain damage incurred is delayed, suggesting that the calcium transient gives rise to sustained changes in membrane function and metabolism. If the ischemia is less dense, as in the penumbral zone of a focal ischemic lesion, pump failure may be moderate and the leak may be only slightly or intermittently enhanced. These differences in the pump/leak relationship for calcium explain why calcium and glutamate antagonists may lack effect on the cardiac arrest type of ischemia, while decreasing infarct size in focal ischemia. The adverse effects of acidosis may be exerted by several mechanisms. When the ischemia is sustained, acidosis may promote edema formation by inducing Na+ and C1- accumulation via coupled Na+/H+ and C1-/ HC03- exchange; however, it may also prevent recovery of mitochondria1 metabolism and resumption of H+ extrusion. If the ischemia is transient, pronounced intraischemic acidosis triggers delayed damage characterized by gross edema and seizures. ~ossibl~~ this is a result of free-radical formation. If the ischemia is moderate, as in the penumbral zone of a focal ischemic lesion, the effect of acidosis is controversial. In fact, enhanced glucolysis may then be beneficial. Although free radicals have long been assumed to be mediators of ischemic cell death, it is only recently that more substantial evidence of their participation has been produced. It now seems likely that one major target of free radicals is the microvasculature, and that free radicals and other mediators of inflammatory reactions (such as platelet-activating factor) aggravate the ischemic lesion by causing microvascular dysfunction and blood-brain barrier disruption. Solid experimental evidence exists that the infarct resulting from middle cerebral artery occlusion can be reduced by glutamate antagonists, by several calcium antagonists, and by some drugs acting on Ca++ and Na+ influx. In addition, published reports hint that qualitatively similar results are obtained with drugs whose sole or main effect is to scavenge free radicals. Thus, there is substantial experimental evidence that the ischemic lesions due to middle cerebral artery occlusion can be ameliorated by drugs, sometimes dramatically; however, the therapeutic window seems small, maximally 3 to 6 hours. This suggests that if these therapeutic principles are to be successfully applied to the clinical situation, patient management must change. KEY WORDS cerebral ischemia penumbra reperfusion brain injury acidosis free radical T HIS is the second part of a review devoted to the ischemia. Part IIs8 discussed the pathophysiol- pathophysiology and treatment of focal cerebral ogy of focal ischemic lesions and the major metabolic events triggered by ischemia. In the present article, the discussion focuses on cellular and molecular mecha- J. Neurosurg. / Volume 77 /September, 1992 337