ESTORATION of cerebral blood flow after prolonged and severe ischemia may cause damage to the blood-brain barrier (BBB), exacerbate brain ede- ma, and cause intracerebral hemorrhage. 2,6,11,21,22,39 Thus, reperfusion injury is a potentially hazardous complication of surgical revascularization, temporary intraoperative cerebrovascular occlusion, or thrombolytic therapy for acute stroke. Recent experimental evidence suggests that arachidon- ic acid and its oxidative metabolites (eicosanoids) may be important mediators of ischemic injury. 26,28,45 Ischemia stimulates the release of arachidonic acid from membrane phospholipids. 13,38 When blood flow and oxygenation are restored, oxidation of arachidonic acid and formation of eicosanoids increase. 10,27,47 These reactions may exacer- bate tissue injury through vasoconstrictive, prothrom- botic, chemoattractant, or direct neurotoxic actions. 4,42,43 The pathways involved in the metabolism of arachidon- ic acid are diagrammed in Fig. 1. Thromboxanes and prostaglandins are produced via the cyclooxygenase pathway. The lipoxygenases produce leukotrienes, lipox- ins, and hepoxillins, as well as hydroxyeicosatetraenoic acid (HETE). Blockage of arachidonic acid metabolism by cyclooxy- genase inhibitors such as aspirin is useful in preventing stroke, but there is little evidence that cyclooxygenase inhibitors are effective in reducing damage after stroke. One explanation is that arachidonic acid may be diverted to the lipoxygenase pathways if cyclooxygenase is inhib- ited; therefore, a more effective approach may be to inhib- it both cyclooxygenase and lipoxygenase pathways. The agent BW755C (3-amino-1-[ m(trifluoromethyl)phenyl]- 2-pyrazoline) (Wellcome Research Laboratory, Kent, England) is a nonsteroidal antiinflammatory agent that inhibits monooxygenases including cyclooxygenase and 12- and 15-lipoxygenase. 20,37 This drug is a rational choice for treating cerebral reperfusion injury because 15-lipoxy- genase and cyclooxygenase metabolites are potent cere- bral vasoconstrictors, 25 and the leukotrienes produced via the 5-lipoxygenase pathway disrupt the BBB. 5 The drug has been shown to reduce infarction in myocardial ischemia–reperfusion models. 33,43 Its central nervous sys- tem activity has been demonstrated by improvement in neurological outcome as well as inhibition of thrombox- ane B 2 production after spinal cord ischemia and trau- ma. 16,17 The action of BW755C may not be limited to J. Neurosurg. / Volume 83 / July, 1995 J Neurosurg 83:99–104, 1995 Attenuation of postischemic brain hypoperfusion and reperfusion injury by the cyclooxygenase–lipoxygenase inhibitor BW755C JUN CHEN, M.D., PHILIP R. WEINSTEIN, M.D., AND STEVEN H. GRAHAM, M.D., PH.D. Departments of Neurology and Neurological Surgery, School of Medicine, University of California, San Francisco; and Department of Veterans Affairs, Veterans Administration Hospital, San Francisco, California  Arachidonic acid metabolites are believed to be important mediators of tissue injury during reperfusion after cere- bral ischemia. To determine whether inhibiting the oxygen-dependent metabolism of arachidonic acid would reduce reperfusion injury, we administered the mixed cyclooxygenase–lipoxygenase inhibitor BW755C (3-amino-1-[ m(tri- fluoromethyl)phenyl]-2-pyrazoline) near the time of reperfusion in a rat model of temporary focal ischemia. The dura- tion of ischemia + reperfusion was 2 hours + 22 hours, 3 hours + 3 hours, or 3 hours + 21 hours. The effects of drug or saline treatment on infarct volume, blood-brain barrier permeability, and blood flow were determined. Cortical blood flow was monitored with laser Doppler flowmetry and blood-brain barrier permeability was evaluated by the Evans blue dye method. Infarct volume was determined in all groups by computerized image analysis of Nissl-stained sections. We found that BW755C treatment significantly attenuated delayed postischemic hypoperfusion in the 3 + 3 group (p  0.05) and reduced the volume of Evans blue dye staining in the cortex (p  0.01) and basal ganglia (p  0.05). Hemispheric swelling was reduced in all treatment groups (p  0.01), as was total infarct volume in the ischemic hemisphere (p  0.05). These results support the hypothesis that arachidonic acid metabolites contribute to acute postischemic reperfusion injury and suggest that using a mixed cyclooxygenase–lipoxygenase inhibitor as an adjunct to thrombolytic or revascularization therapy could lengthen the ischemia time after which reperfusion is beneficial. KEY WORDS • cerebral ischemia • reperfusion • permeability • infarction • eicosanoids • cerebral blood flow R 99