[The application of intravoxel incoherent motion diffusion imaging and 3D arterial spin labeling perfusion imaging in the evaluation of acute cerebral infarction].

To investigate the clinical value of magnetic resonance (MR) intravoxel incoherent motion (IVIM) diffusion imaging and 3D pseudo continuous arterial spin labeling (3D-pcASL) perfusion imaging in the evaluation of acute cerebral infarction. MR images of 49 patients with unilateral acute cerebral infarction diagnosed and treated in Affiliated Yancheng Hospital of Southeast University Medical College from October 2015 to February 2019 were retrospectively analyzed. High signal infarction area (S(D)) on diffusion image slice with the biggest lesion level and abnormal perfusion area (S(CBF)) on the corresponding level were measured. The presence of ischemic penumbra (IP) was represented by S(CBF) S(D), and patients were divided into group IP and group non-IP. Regions of interest were set on the infarction core, brain tissue near the edge of the lesion (BNL) and their corresponding contralateral regions. The values of apparent diffusion coefficient (ADC), true diffusion coefficient (D), perfusion related diffusion coefficient (D()), perfusion fraction (f) and cerebral blood flow (CBF) of each region of interest were recorded and relative values of infarction lesion to its contralateral region (rADC, rD, rD(), rf, rCBF) were calculated. Differences of each parameter value between infarction core, BNL and their corresponding contralateral regions and of each relative parameter value between infarction core and BNL, and between the two groups were compared.The differential diagnostic efficacy of relative parameter value with differences between groups was analyzed by receiver operating characteristics (ROC) curve. The correlations of each relative parameter value of 3D-pcASL and IVIM sequences were analyzed. The ADC, D, f and CBF values of infarction core were significantly lower than those of contralateral regions in both groups (all 0.01). Among all parameters of BNL in both groups, only the CBF value of group IP was significantly lower than that of contralateral region ((27.58±3.53) vs (41.20±5.66) ml·100 g(-1)·min(-1), 0.01). The rADC, rD, rf and rCBF of infarction core were significantly lower than those of BNL in both groups (all 0.01). The rCBF of BNL in group IP was significantly lower than that in group non-IP (0.68±0.12 vs 0.97±0.15, 0.01), and the area under the curve was 0.949, the optimal threshold was 0.823, and the youden index was 0.855 for identifying the two groups. Other relative parameters values of infarction core and BNL had no statistical difference between the two groups. There were positive correlations between rCBF and rADC, rD, rf (0.428,0.335,0.565) of infarction core, rADC and rD, rf (0.853,0.602) of infarction core, also rADC and rD (0.336) of BNL (all 0.05). IVIM can effectively evaluate the difusion and perfusion information of acute cerebral infarction lesions. However, its perfusion related parameters are not as good as 3D-pcASL in IP evaluation, which should be flexibly selected according to the actual needs of patients' condition evaluation.