Background

Functional magnetic resonance imaging (fMRI) has been used to localize memory related blood oxygen level dependent (BOLD) networks and predict post-operative memory outcome in patients undergoing epilepsy surgery. However, delineation of these networks is largely dependent on the memory paradigm used during fMRI experiments and has its limitations. Hometown walking paradigm (HWP) focuses on retrieval of stored information and shown to activate medial temporal structures reliably.

Objective

Our aim was to use home town walking paradigm in clinical settings to localize memory function in patients undergoing pre-surgical evaluation.

Methods

Patients were selected from MDT from Jan 2015 to April 2022. All patients had clinical assessment, video-telemetry and structural imaging . Pts. had detailed neuropsychology assessments for memory functions before and after surgery. We used HWP for memory localisation which required mental navigation through one’s hometown by using landmarks given by patients themselves. We performed fMRI acquisitions on two consecutive days. fMRI data was acquired on a Siemens 3T scanner, using block design in General Linear Model and analysed using manufacturer BOLD processing package . The temporal activation t-maps are thresholded for minimum allowable t-value and fused to 3D T1W images using Syngo VIA software.

Results

244 fMRI studies were performed. Majority of patients showed BOLD activations in medial temporal lobes, however there were around 8% of patients who did not show BOLD activations despite repeating the fMRI study on day 2. Stronger BOLD activations were seen in parahippocampal and fusiform gyri. Bilateral activations were seen in parahippocampal gyri in most of patients, and patients with lesions in temporal lobes had stronger activation in contralateral parahippocampal gyrus. Additional activations were seen in posterior cingulate, parietal cortex and dorsolateral prefrontal cortex reflecting visuospatial mental navigation and sequencing of task during hometown walk. In patients who proceeded to surgical resection, none reported a new cognitive memory deficit subjectively. Post-operative assessments showed improvement of auditory and visual memory index as compared to preoperative state in two cases.

Conclusions

Implementation of HW paradigm show consistent and reproducible BOLD activations in medial temporal lobes to localise memory in pts. undergoing pre-surgical evaluation, which can be used as an adjunct to neuropsychological memory assessment.