Abstract:Objective To analyze the magnetic resonance arterial spin labeling (ASL) in grading of brain glioma. Methods A retrospective analysis of 84 patients with glioma confirmed by surgery and pathology in the Department of Neurosurgery of our hospital from January 2015 to January 2017 was made. All patients received axial T1WI, T2WI, T2 FLAIR, DWI and 3D-ASL scans, and then received enhanced scan of T1WI. The 3D ASL scan with pseudo continuous artery self rotation labeling technique was used to calculate cerebral blood flow (CBF) value and relative cerebral blood flow (rCBF) was then transformed. The conventional sequence and 3D ASL perfusion imaging were analyzed. Results A total of 84 patients received conventional MRI image sequences and 3D ASLCBF perfusion map with 3DASL-CBF perfusion map, glioma patients were as follows: red area represented high blood perfusion; blue area represented low perfusion. The ratios of TBF/contralateral white matter, TBF/contralateral gray matter and TBF/contralateral hemisphere in grade I, II and III, IV gliomas were statistically significant (P <0.05). Pathological examination showed that there were 45 cases of advanced glioma and 39 cases of low-grade glioma. Compared with pathological results, the diagnostic accuracy of 3D ASL was 92.86% (78/84), which was higher than that of conventional MRI 82.14% (69/84) (P < 0.05). Kappa value of table was 0.758, the consistency of which was good. Conclusions 3D ASL can quantitatively reflect the perfusion of tumor microcirculation. Compared with routine MRI sequence, it is of important reference value for qualitative diagnosis and preoperative grading of glioma.