Usefulness of automated tractography for outcome prediction in patients with recurrent stroke.

[Purpose] To examine the usefulness of automated tractography for predicting outcomes in patients with recurrent stroke. [Participants and Methods] Diffusion tensor imaging was performed in the second week after stroke, and fractional anisotropy was calculated using automated tractography. Three patients with recurrent strokes were included in this study. [Results] Initial computed tomography findings of a 62-year-old man with stuttering speech revealed a hemorrhage in the left thalamus. Fractional anisotropy indicated slight neural damage in the association fibers of both hemispheres. The patient returned to work with mild attention deficit and aphasia. Initial diffusion-weighted imaging of a 75-year-old man with right upper extremity paresis showed high-intensity areas in the left corona radiata. Fractional anisotropy indicated bilateral neural damage to the corticospinal tract. The patient was discharged with severe right upper extremity impairment and a modified gait. Initial diffusion-weighted imaging of a 60-year-old woman with moyamoya disease who experienced a sudden loss of consciousness showed high-intensity areas in the left anterior circulation territories. Fractional anisotropy indicated severe damage to the right hemisphere, the corticospinal tract, and the superior longitudinal fasciculus of the left hemisphere. She was transferred to a nursing home and remained bedridden. [Conclusion] The symptoms identified in this study agreed with automated tractography findings, which suggests that this methodology is useful for predicting recurrent stroke outcomes.