Reperfusion and reperfusion injury in cerebral ischemia in cats as studied by invasive and non-invasive techniques. The role of free radicals, neurotoxins and edema

Abstract

Although the clinical features of space-occupying ischemic stroke are well known, there are limited data on the clinical course of complete middle cerebral artery territory (MCA) infarction and on the predisposing factors leading to subsequent herniation and brain death. One major reason is, that pathophysiological processes in the human brain are heterogenous both in the temporal and spatial domain. Following occlusion of the left MCA in halothane anesthesized cats, extracellular amino acid and purine catabolite concentrations and local cerebral blood flow (CBF) were repeatedly monitored during 3h of occlusion and 6h of reperfusion in core and perifocal cortical regions of the ischemic focus using microdialysis (MD) and Laser Doppler (LD) probes. MCA occlusion reduced LDF-CBF in all animals to below 25% of control, and increased extracellular glutamate about 15 times in core but not in perifocal regions. In the ischemic focus, aspartate, γ-aminobutyrate (GABA) as well as hypoxanthine, xanthine and inosine reached maximum levels 1-2h after onset of ischemia in those animals that developed malignant edema. The elevation of hypoxanthine, inosine and xanthine persisted during ischemia, whereas that of adenosine was only transient, lasting - 1h. Upon reperfusion, LDF-CBF and glutamate primarily recovered. In 5 of 10 cats, secondary elevation of glutamate and other neurochemicals was apparent during the reperfusion phase in both core and perifocal regions. These started to increase when cerebral perfusion pressure (CPP) decreased below 60 mmHg. At the same time of secondary glutamate elevation, symptoms of transtentorial herniation were recognised. Histology revealed early neuronal necrosis progressing to the contralateral cortices. In the other five cats, secondary elevation of glutamate and other neurochemicals was not recognised during the reperfusion stage, and malignant edema did not result. The use of the salicylate trapping technique combined with MD, it was shown that the absolute amount of hydroxyl radicals produced in the ischemic and early reperfusion phase in the malignant group, accompanied the larger infarct volume, more severe edema and greater perfusion deficit than that found in the non-malignant group. The higher levels of hydroxyl radicals found in the perifocal region of both groups might be related to the sustained substrate elevation of purine catabolites that are known to participate in free radical mechanisms. Conversely, it is suggested that formation of hydroxyl radical production in the ischemic core involves a different mechanism than in perifocal regions, and that augmented excitatory and free radical dynamics during times when metabolic states are already compromised contribute to a wider vicious cycle. Sequential PET measurements revealed a variable and often unpredictable evolution of ischemic and post-ischemic CBP with transitions from hyper- to hypoperfusion during recirculation. Post-ischemic reperfusion patterns, however, seemed to be more variable than the initial ischemic disturbances. In the cats that developed malignancy, extended periods of post-ischemic hyperperfusion were accompanied by large defects in the cerebral metabolic rate of oxygen (CMR02), the development of large infarcts and a fatal increase in intracranial pressure (ICP). On the other hand, acute-stage transient postischemic hyperperfusion unassociated with extensive severe hypometabolism did not produce vasogenic edema with the accompanied progression into secondary deterioration. In complementary experiments, 7 of 9 cats showed an increase in NO during ischemia in the ischemic core. During reperfusion, 6 of 9 cats showed a variable increase in NO. Considering the different patterns of NO production in core and perinfarct regions, it would appear plausible that local regional differences in NOS activity and free radical generation contribute to either the severity or progression of injury. In conclusion, good correlations exist between malignant edema formation and secondary elevations of substances that may be involved in processes of delayed deterioration. Prediction of this fatal progression, however, may be achieved better by modern imaging techniques such as PET or MRI that provide views of the whole brain and thus permit judgement on a broader basis. Thus, the continuous intensive neuromonitoring of local parameters using invasive tools may be of supplementary benefit for the prognosis and treatment of malignant, space-occupying edema in severe stroke.