Mean apparent propagator-MRI: A new diffusion model which improves temporal lobe epilepsy lateralization.

PURPOSE

To compare MRI volume measurements, FLAIR image intensity, Diffusion tensor imaging (DTI) and mean apparent propagator (MAP)-MRI measurements in hippocampus ipsilateral and contralateral to the epileptogenic focus for non-invasive lateralization of temporal lobe epilepsy (TLE) and also compare these DTI and MAP-MRI measurements to cognitive function.

METHOD

A cohort of patients with unilateral TLE and aged-and gendered-matched controls were enrolled in this retrospective study. T1-weighted MPRAGE data for the volume, FLAIR image intensity, DTI and MAP-MRI parameters were performed for bilateral hippocampi of all subjects. The sensitivity, specificity, lateralization ratios and Cohen's d effect sizes of all MR measurements were calculated. Pearson correlation analysis was performed to compare DTI and MAP-MRI measurements to cognitive function.

RESULTS

We evaluated 23 patients and 17 controls. The MAP-MRI parameter 'return to the plane probability' (RTPP) had the strongest effect size (d = -1.678, lateralization ratio = 86.36 %) for differentiating hippocampus ipsilateral to the epileptogenic focus from contralateral hippocampus when compared to all other DTI/MAP-MRI parameters, signal intensity on FLAIR and hippocampal volumes. Mean diffusivity (MD), radial diffusivity (RD), mean square displacement (MSD) were each negatively correlated to clinical measures of delayed recall (r = -0.758; r = -0.772; r = -0.684, respectively). While return to the axis probability (RTAP) return to the origin probability (RTOP) and fractional anisotropy (FA) were positively correlated (r = 0.832; r = 0.813; r = 0.717, respectively) (all P < 0.05).

CONCLUSION

MAP-MRI measurements are promising radiologic biomarkers for the non-invasive lateralization of epileptogenic foci in TLE.