A study of neurite orientation dispersion and density imaging in ischemic stroke.

OBJECTIVES

To demonstrate the feasibility of the neurite orientation dispersion and density imaging (NODDI) technique in characterizing the microstructural changes in brain tissues during ischemic stroke and to compare its sensitivity with diffusion tensor imaging (DTI) and diffusion kurtosis imaging (DKI).

METHODS

Seventy-one patients with hyperacute/acute/subacute ischemic stroke were enrolled in the study. A multishell diffusion magnetic resonance imaging (dMRI) protocol was performed for each subject. Diffusion data were analyzed using the NODDI and diffusional kurtosis estimator toolboxes. Then, NODDI metrics between the lesions and the contralateral tissues were compared to evaluate their values in ischemic stroke. NODDI metrics among different stroke periods and the correlations between NODDI and the duration since stroke onset were analyzed as well. To compare the NODDI's sensitivity with established diffusion techniques, paired t-tests were performed to determine the absolute percentage changes of diffusion metrics between NODDI and DTI/DKI.

RESULTS

Compared with the contralateral tissues, lesions showed significantly increased values of intracellular volume fraction (V) and orientation dispersion index (ODI) and decreased values of isotropic volume fraction (V). ODI value was significantly different among three periods and showed fair to good positive correlation with the duration since stroke onset (R = 0.450). NODDI metrics showed significantly larger absolute percentage changes than that of DTI and DKI (P < 0.05, respectively).

CONCLUSION

NODDI allowed efficient evaluation of microstructural changes in brain tissues during ischemic stroke and showed increased sensitivity compared with DTI and DKI. The possible biophysical mechanisms underlying ischemia could be further elucidated using this advanced diffusion technique.