Journal of International Psychiatry

Objective To explore application value of diffusion tensor imaging（DTI）result for predicting early post stroke cognitive impairment（PSCI）.Methods Retrospective analysis were performed on 62 early stroke combined with PSCI patients（PSCI group）treated in our hospital from December 2018 to December 2019，and 60 healthy controls were employed as control group.DTI results and cognitive functions scores by Montreal Cognitive Assessment（MoCA）were primary observed indicators.Two groups’ DTI results were comparative analysed，analysing the correlation between fractional anisotropy（FA）and MoCA scores，constructing ROC curve to assess prediction efficiency of DTI for predicting PSCI.Result Bilateral frontal lobe FA，bilateral hippocampus FA，callosum FA in PSCI group were lower than those in the control group（P<0.05）；Bilateral frontal lobe FA（left：r=0.499，P=0.000；right：r=0.395，P=0.000），bilateral hippocampus FA（left：r=0.355，P=0.000；right：r=0.506，P=0.000），callosum FA（r=0.409，P=0.000）were positive correlation with MoCA scores；Cut-off of right hippocampus FA to predict early PSCI was 0.22，the AUC was 0.911 and higher than that of left hippocampus，right frontal lobe，callosum（Z=2.488，2.039，2.936；P=0.013,0.041,0.003），the sensitivity of right hippocampus FA，left hippocampus FA，left frontal lobe FA were 93.50%，85.52%，83.92% and were better than other indicators（P<0.05）.Conclusion DTI can show the abnormal connection in brain networks in early PSCI patients，FA variation on brain white matter of bilateral frontal lobe，bilateral hippocampus and callosum have a good predictive performance for PSCI.

PSCI；hippocampus；MoCA