Experimental Research On Superacute Cerebral Ischemia-reperfusion Injury With Magnetic Resonance Imaging

Background and purpose: Cerebral infarct is a universal disease threatening human health, and thrombolytic therapy is an important method for saving ischemic cerebral tissue. But animal experiments and clinical studies indicated that there was unavoidable reperfusion injury associated with blood flow restored in ischemic cerebral tissue. Therefore, it is of great importance to dynamically monitor the cerebral ischemia-reperfusion (I/R) injury with sensitive and noninvasive method in medical field, so as to provide timely pathophysiological information for its clinical prevention and treatment. With the advance of new MRI techniques recently, especially the development of diffusion-weighted imaging (DWI) and perfusion-weighted imaging (PWI) techniques, it provided an advanced method for the monitor of cerebral I/R injury. The aim of our study was to monitor the time-courses of cerebral I/R injury in superacute transient middle cerebral artery occlusion (MCAO) models, which were copied with a modified suture occlusion method in rats, by dynamic MRI scans. Furthermore, influence of different reperfusion time-points on the evolvement of the ischemic cerebral tissue were explored combined with its pathological examination, and the value of MRI in monitoring superacute cerebral I/R injury were evaluated, to provide theoretical foundation and experimental support for clinical application.Materials and methods: 36 healthy male Wistar rats, admitted by the Experimental Animal Center of Henan Province, were randomly and equally divided into 6 groups, weight of rats among 240~280g. In control group (A), tip of the suture was just inserted within the left external carotid artery (ECA). For permanent MCAO group (B), the left MCA was occluded with the suture. As to reperfusion groups (C, D and E), tip of the suture waswithdrawn into the ECA at 30 minutes, lh and 2h after MCAO respectively. In pathology group (F), one model corresponding to each previous group was copied, cerebral specimens of these models acquired at 6h after operation. An additional model corresponding to group C was made for getting specimen at lh after reperfusion.Coronal sections were performed with Marconi Eclipse 1.5T MR imager on the rat brain. Within 6h after MCAO, DWI was performed at each 30 minutes interval, PWI, T2WI and enhanced TiWI were performed at each lh interval for group A and B. For group C, D and E, DWI was performed at pre-reperfusion and each 30 minutes after reperfusion within 6h after MCAO. In addition, PWI, T2WI and enhanced TiWI were performed at pre-reperfusion and each lh after MCAO for group C, at pre-reperfusion and each lh after reperfusion for group D and E. Specimens for group F were gained by cutting brains corresponding to the slice of MRI scan, observed under light and electron microscope respectively.Image reconstruction was accomplished automatically with the associated software. On the slice which included bilateral lateral ventricles, abnormal signal area on DWI, hemisphere area, signal intensity of the region of interest (ROI) on TiWI and T2WI, and apparent diffusion coefficient (ADC) value on ADC maps were measured. The relative high signal area on DWI (rSowi), relative signal intensity of the ROI on TiWI (TiWI ROI rSI) and T2WI (T2WI ROI rSI), and relative ADC (rADC) value were calculated for the operation side. Data were expressed as x±s, and were analyzed with SPSS 10.0. Comparisons between two groups and among multiple groups were performed with the use of Mest and one-way ANOVA respectively. Correlation between two variables was expressed with Pearson coefficient. Significant level was set as a =0.05.Results: (1) It was normal on MRI at every time-point in group A. (2)In group B, relative high signal on ischemic side was found all the time on PWI, and the absolute peak value of signal intensity versus time curve (SI-TC) was less than that of the contralateral side. High signal was found 30 minutes after MCAO on DWI, and the ADC value on the corresponding area declined. Abnormal signal area increased with time on DWI, and the ADC value decreased further. High signal initially appeared on T2WI 2h after MCAO, and enhancement was initially found on TiWI 3h after MCAO. Signal intensity and extent on both and T2WI increased with time. (3) For group C, relative high signal reduced sharplyafter reperfusion on PWI, the peak value of SI-TC approached that of the contralateral side. Abnormal signal extent reduced prominently and the ADC value increased back rapidly after reperfusion, but there is a second decrease for ADC values lh after reperfusion. No obvious abnormal signal was found on either TiWI or T2WI all along. (4) In group D, relative high signal reduced sharply after reperfusion on PWI, the peak value of SI-TC approached that of the contralateral side. Abnormal signal extent reduced prominently and the ADC value in 3 models increased back rapidly after reperfusion, but there was a second decrease for ADC values lh after reperfusion. In other 3 models, the ADC value kept decreasing slowly. High signal initially occurred 3h after MCAO on T2WI and enhancement initially was found 4h after MCAO on TiWI in 5 models. No abnormal signal was found in 1 model. (5) As to group E, relatively high signal intensity reduced slowly after reperfusion on PWI, the absolute peak value of SI-TC is less than that of the contralateral side. There was no obvious change of the abnormal signal extent on DWI, and ADC value remained decreasing slowly. High signal could be found instantly after reperfusion on T2WI, enhancement could be visualized 3h after reperfusion on TiWI, the signal intensity and extent on both TiWI and T2WI increased with time. (6) Value of rSowi was zero at each time-point in group A, increased slowly with time in group B, decreased sharply in group C, decreased slowly in group D and no obvious change after reperfusion in group E. Value of rADC was approaching zero at each time-point in group A, decreased continuously with time in group B, increased sharply after reperfusion, but with a second decrease lh later in group C. For group D, there was a mild increase of the rADC after reperfusion, but with a second decrease lh later. In group E, rADC decreased slowly after reperfusion. TiWI ROI rSI of cortex and striatum was close to null at each time-point in group A, increased sharply at 5h after operation in group B, increased slowly with time after reperfusion in group C, initially increased sharply at 6h and 5h after MCAO respectively on cortex and striatum for group D, initially increased sharply at 5h after MCAO in group E. T2WI ROI rSI of cortex and striatum was close to null at each time-point in group A, initially increased sharply at 2h after operation in group B, increased slowly with time after reperfusion in group C, initially increased sharply at 3h and 2h after MCAO on cortex and striatum respectively for group D, increased sharply soon after reperfusion in group E. (7) In sham model, there were all kinds of normal organelles within the neuron, which remainedregular shape. As to permanent MCAO model, the neuron was swollen with nuclear concentration and mitochondrial balloon-like swelling. There was a enlarged transparent strip around the neuron. In the model of reperfusion after 30 minutes MCAO, neuronal shape was clear with mitochondrial slightly swelling and a slightly narrow transparent strip around the neuron lh after reperfusion, while with obvious mitochondrial swelling and mild extracellular edema at 5h 30minutes after reperfusion. In the model of reperfusion after lh MCAO, the mitochondrial was swelling with cytoplasmic and extracellular edema. In the model of reperfusion after 2h MCAO, there was very obvious mitochondrial swollen and cytoplasmic edema. The transparent space around the neuron was widden. (8) On the same model, the occurrence of high signal on T2WI was ahead of enhancement appearance on TiWI, and the spacial distributions of abnormal signal coincide with each other on TiWI and T2WI. There was neither obvious correlation between TiWI ROI rSI and T2WI ROI rSI on cortex (r=0.116, P=0.463) nor on striatum (r=-0.006, P=0.971) before T}WI ROI rSI initially increased with statistical significance, though T2WI ROI rSI did. There was a positive correlation between TiWI ROI rSI and T2WI ROI rSI on cortex (r=0.459, P=0.011) and striatum (r=0.566, /^O.OOO) respectively after TXWI ROI rSI and T2WI ROI rSI both increased with statistical significance.Conclusion: (1) There was still reperfusion injury associated with reperfusion in superacute cerebral ischemia. The extent and degree of reperfusion injury was related to the ischemic duration, the longer the ischemic duration the severe reperfusion injury might occur. (2) Reperfusion injury was unavoidable, but it was feasible and indispensable to save reversible ischemic tissue by restoring its blood flow, earlier its blood flow restored more reversible ischemic tissue might be saved. (3) There was a positive correlation between vasogenic edema and increasing of BBB permeability after cerebral ischemia reperfusion. However, contrast enhanced TiWI showed lack of sensitivity to early slight increase of BBB permeability. (4) MRI had proved to be an ideal and uninvasive modality to monitor cerebral reperfusion injury. More information could be acquired by various MRI techniques combined with each other.