Time-Dependent Functional Specialization of Hippocampal Subfields Detected by MRI in Patients with Temporal Lobe Epilepsy
Abstract number :
1.377
Submission category :
11. Behavior/Neuropsychology/Language / 11A. Adult
Year :
2019
Submission ID :
2421370
Source :
www.aesnet.org
Presentation date :
12/7/2019 6:00:00 PM
Published date :
Nov 25, 2019, 12:14 PM
Authors :
Takahiko Mukaino, Kyushu University; Taira Uehara, Kyushu University; Jun Yokoyama, Kyushu university; Toshiki Okadome, Kyushu University; Ayumi Sakata, Kyushu University Hospital; Tomomi Arakawa, Kyushu University Hospital; Setsu Yokoyama, Fukuoka Sanno
Rationale: Patients with temporal lobe epilepsy (TLE) have been shown to exhibit memory deficits, including the rapid forgetting of newly acquired material over long delays (termed accelerated long-term forgetting; ALF). We have started a clinical trial to clarify the precipitating factors for ALF. Using these data, we investigated the functional specialization of hippocampal subfields detected by MRI in short- and long-term memory. Methods: TLE patients (n=65) with and without hippocampal sclerosis and healthy controls (HC) (n=30) were enrolled from our ongoing clinical trial. All participants underwent background neuropsychological assessment, including simplified Wechsler Adult Intelligence Scale (WAIS-Ⅲ), Standardized Paired Associates (S-PA), and Benton visual retention test. Using two novel experimental tasks to measure long-term forgetting, cued recall of verbal (story) and visuospatial (route) materials were tested 30 secs, 10 mins and 1 week after learning. All TLE patients underwent 3 dimensional T1 weighted image (WI) and coronal T2WI with 3 Tesla MRI. Hippocampal subfields were automatically segmented with FreeSurfer software. To assess the presence of ALF, group and time interaction were assessed with mixed ANOVAs between HC and TLE. In TLE, the relationship between the volume of each hippocampal subfield and task performance at each time interval was assessed by multiple regression analysis, in which age, medication, intelligent quotient, known left-sided seizure activity, and seizure occurrence after learning were adjusted. Results: The two-way group*delay interval ANOVA showed significant main group effects with both story and route tasks, but group-by-time interactions were not significant. In multiple regression analysis, right whole hippocampus (p = 0.02) and its internal structural volumes (presubiculum, CA1, molecular layer, granular cell of dentate gyrus, CA2/3, and CA4) (p < 0.05 for each) were associated with 30 secs delayed recall score in story task. Right presubiculum (p = 0.03) and left CA1 (p = 0.04) volumes predicted 10 mins delayed recall score, and only left CA1 (p = 0.04) volume predicted 1 week delayed recall. With route task, right amygdala (p = 0.01) volume was associated with 10 mins delayed recall score. Conclusions: The right whole hippocampal volume rather than specific hippocampal subfields was associated with short-term verbal memory. However, only left CA1 was associated with 1 week delayed verbal memory performance. Based on these findings, we speculate that CA1 is involved in long-term memory consolidation, ultimately ALF in TLE patients. Funding: MEXT KAKENHI Grant Number JP17K09802
Behavior/Neuropsychology/Language