Premium
Resting‐state functional MRI distinguishes temporal lobe epilepsy subtypes
Author(s) -
Reyes Anny,
Thesen Thomas,
Wang Xiuyuan,
Hahn Daniel,
Yoo Daeil,
Kuzniecky Ruben,
Devinsky Orrin,
Blackmon Karen
Publication year - 2016
Publication title -
epilepsia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.687
H-Index - 191
eISSN - 1528-1167
pISSN - 0013-9580
DOI - 10.1111/epi.13456
Subject(s) - temporal lobe , hippocampal sclerosis , epilepsy , hippocampus , amygdala , medicine , magnetic resonance imaging , psychology , nuclear medicine , neuroscience , audiology , radiology
Summary Objective We assessed whether presurgical resting state functional magnetic resonance imaging (fMRI) provides information for distinguishing temporal lobe epilepsy (TLE) with mesial temporal sclerosis (TLE‐MTS) from TLE without MTS (TLE‐noMTS). Methods Thirty‐four patients with TLE and 34 sex‐/age‐matched controls consented to a research imaging protocol. MTS status was confirmed by histologic evaluation of surgical tissue (TLE‐MTS = 16; TLE‐noMTS = 18). The fractional amplitude of low‐frequency fluctuations (fALFFs) in the blood oxygen level–dependent (BOLD) resting‐state fMRI signal, a marker of local metabolic demand at rest, was averaged at five regions of interest (ROIs; hippocampus, amygdala, frontal, occipital, and temporal lobe), along with corresponding volume and cortical thickness estimates. ROIs were labeled ipsilateral or contralateral according to seizure lateralization and compared across TLE‐MTS, TLE‐noMTS, and healthy controls (HCs). MTS status was regressed on ipsilateral hippocampal volume and fALFF to test for independent contributions. Results The TLE‐MTS group had reduced fALFF in the ipsilateral amygdala and hippocampus; whereas, the TLE‐noMTS group had marginally reduced fALFF in the ipsilateral amygdala but not hippocampus. These results were consistently obtained with and without application of global signal regression (GSR). Ipsilateral hippocampal volume contributed to 37% of the variance in MTS status (p < 0.001) and fALFF contributed an additional 10% (p = 0.021). Two MTS cases were accurately classified with fALFF but not volume, and three were accurately classified with volume but not fALFF. At the lobar level, fALFF (with GSR) was reduced in the ipsilateral temporal and bilateral frontal lobes of patients with TLE‐MTS and bilateral frontal lobes of patients with TLE‐noMTS in the context of normal cortical thickness. Significance This study indicates that resting‐state fMRI provides complementary functional information for MTS classification. Findings validate fALFF as a measure of regional brain integrity in TLE and highlight the value of using multi‐modal imaging to provide independent diagnostic information in presurgical epilepsy evaluations.