Objective:

To use magnetic resonance fingerprinting (MRF) to identify T1 and T2 tissue property differences in hippocampal subfields of elderly patients with epilepsy.

Background:

Elderly patients with epilepsy have increased rates of impaired cognition. MRF is a novel imaging modality that allows for fast and quantitative measurement of intrinsic tissue properties on a voxel-by-voxel basis, which can help identify structural biomarkers of impaired cognition. In this study, we demonstrate the utility of high-resolution 3D MRF in identifying structural differences within hippocampal subfields in elderly patients with epilepsy. 

Design/Methods:

Patients 55 years or older with non-lesional focal epilepsy were included. Participants underwent whole-brain 3T-MRF and conventional MRI using a protocol adapted from the Alzheimer’s Disease Neuroimaging Initiative. After reconstruction of raw MRF data, we generated quantitative T1 and T2 relaxometry maps and co-registered them with T1-weighted and T2-weighted conventional MRI images using Advanced Normalization Tools. We then utilized FreeSurfer and the FS60 atlas to segment hippocampal subfields. T1 and T2 relaxation times in each subfield were then calculated. As a pilot study, we compared the T1 and T2 values for the ipsilateral and contralateral hippocampal subfields for each patient. 

Results:

Fifteen patients (mean age 67y, 47% women) with non-lesional conventional MRI were included. Eight patients had unilateral temporal/frontotemporal lobe epilepsy, three had unilateral frontal lobe epilepsy, and four had multifocal bilateral epilepsy. In patients with unilateral frontotemporal/temporal lobe epilepsy, T2 values were higher in the ipsilateral CA3 and presubiculum compared to the contralateral side, with mean percent increases of 10.7±13.6% (p=0.049) and 9.1±5.9% (p=0.002), respectively.  

Conclusions:

MRF can quantitatively discriminate between structural differences in hippocampal subfields in elderly patients with non-lesional temporal lobe epilepsy. In the future, we aim to demonstrate a correlation between subfield differences and cognitive measures, which may help to identify markers of impaired cognition in this patient population.