Abstracts

Rationale: The mechanism of disruption of language networks is poorly understood in temporal lobe epilepsy (TLE) patients. Neuropsychological testing and task-based functional magnetic resonance imaging (fMRI) have been shown to be excellent noninvasive alternatives for presurgical determination of language. However, their usefulness and reliability in measuring hemispheric language dominance can be limited in some patients due to age, language barriers, or states of altered consciousness due to seizures or antiepileptic medications. Thus, task-free approaches such as resting state fMRI (rs-fMRI) may be an overall more reliable tool to evaluate language in these situations. The Amplitude of Low Frequency Fluctuation (ALFF) and fractional Amplitude of Low Frequency Fluctuations (fALFF) are mathematical parameters that measure low fluctuations of activity in the resting brain. We sought to compare ALFF and fALFF within the language networks of patients with medically refractory TLE and similarly aged healthy controls. Methods: Patients undergoing presurgical evaluation at our institution between 2014 and 2019 were retrospectively reviewed. Inclusion criteria included: 1) age > 18 years, 2) unilateral medically refractory TLE confirmed on cv-EEG; 3) completion of rs-fMRI and neuropsychological testing. A total of 15 TLE patients met these criteria. 24 similarly-aged healthy control subjects (HCs) were also included. All patients and HCs underwent rs-fMRI on a 3T Trio-Tim Siemens scanner. rs-fMRI data were preprocessed using SPM (Version 8; Functional Imaging Laboratory, 2009), and ALFF and fALFF were estimated using RESTplus (Version 1.2; Jia et al. 2016). Broca’s and Wernicke’s area were labeled and defined spatially using Automated Anatomical Labeling in SPM. Broca’s area included the inferior frontal gyrus triangularis (IFG-t) and opercularis (IFG-o). Wernicke’s area included the superior and middle temporal gyri (STG and MTG), superior marginal gyrus (SMG), and angular gyrus. Region of interest (ROI) and statistic difference map analyses were utilized and two sample t-tests were performed to compare the ALFF and fALFF differences between TLE patients and HCs. Results: Using an ROI analysis, left TLE patients showed significantly reduced ALFF in right IFG-o and significantly reduced fALFF in bilateral IFG-t and left IFG-o (p<0.001). Using a statistic difference map, left TLE patients showed significantly reduced ALFF in the left SMG and MTG & right IFG (p<0.001). Conclusions: Resting state language networks were significantly disrupted in patients with left TLE. Broca’s area is more affected than Wernicke’s area in both the ipsilateral and contralateral hemispheres. rs-fMRI may be a useful alternative to assess language networks in medically refractory epilepsy. Funding: No funding