[Modeling of acute thromboembolic stroke in mice]

To evaluate the controllability and reproducibility of a acute thromboembolic cerebral ischemia model for studying molecular imaging and thrombolysis in mice by 7.0 Tesla magnetic resonance imaging (MRI).

Twenty-four male C57BL/6J mice were randomly divided into embolic group (n = 14) and sham-operated group (n = 10). To prepare rich fibrin and optimal length of clots and measure diameters of clots under microscope, the clots were injected into internal carotid artery via a microcatheter inserted from external carotid artery to bifurcation of common carotid artery in embolic group. In sham-operated group, phosphate-buffered saline containing bovine serum albumin was injected similarly. At 1, 3 or 24 h after injection of clots or phosphate-buffered saline containing bovine serum albumin, mice underwent 7.0 Tesla magnetic resonance imaging including sequences of magnetic resonance angiography (MRA), diffusion weighted imaging (DWI) and perfusion weighted imaging (PWI) by arterial spin labeling technology (ASL) and T(2) weighted imaging (T(2)WI) so as to evaluate the occlusive rate of middle cerebral artery (MCA) and the change of relative cerebral blood flow (rCBF) and lesion volumes. Triphenyl tetrazolium chloride (TTC) stain was performed at 24 h.

The diameters of clots were 162 ± 14 µm. In embolic group at 1, 3 or 24 h after injection of clots, the occlusive rate of MCA was 78.6%, 71.4% and 57.1% respectively. And the values of rCBF decrease in mice showing occlusion of MCA in MRA images. The percents of rCBF value in embolic group were 26% ± 10% at 1 h and 26% ± 15% at 3 h respectively. They were significantly lower than the percents of rCBF value in sham-operated group at the same time points. The percents of infarct volumes at 24 h were 30% ± 4% from T(2)WI images and 30% ± 16% from TTC images. In sham-operated groups, both MRI and TTC images were negative.

A murine model of acute thromboembolic cerebral ischemia has been successfully established. The improved method is both stable and feasible. It may be readily evaluated from multi-parameter imaging by 7.0 Tesla magnetic resonance imaging.