INS NYC 2024 Program

Poster	Poster Session 09 Program Schedule 02/16/2024 03:30 pm - 04:45 pm Room: Majestic Complex (Posters 61-120) Poster Session 09: Epilepsy \| Oncology \| MS \| Infectious Disease Final Abstract #112 Dentate gyrus volume, but not other medial temporal lobe subregions, predict performance on select neuropsychological measures. Sean Turner, University of Florida, GAINESVILLE, United States Russell Bauer, University of Florida, Gainesville, United States Category: Epilepsy/Seizures Keyword 1: epilepsy / seizure disorders Keyword 2: neuropsychological assessment Keyword 3: hippocampus Objective: Neuropsychological (NP) assessment of surgical epilepsy patients plays a crucial role in localizing areas of dysfunction and predicting postoperative cognitive change. While the specialization of the dominant temporal lobe for language and verbal memory has long been established, considerable controversy exists about the memory functions of the language nondominant temporal lobe, since evidence of material-specific nonverbal memory impairment in right temporal lobe epilepsy (TLE) patients is inconsistent. One lingering question is whether NP test results would reflect with equivalent sensitivity the underlying pathophysiology responsible for language-dominant v. nondominant TLE. Here we sought to investigate whether hippocampal and extra hippocampal subfield volume determined from clinical MRI scans predicted lateralized neuropsychological test performance in a population with temporal-lobe epilepsy. Participants and Methods: Five patients with left-hemisphere TLE and eight patients with right-hemisphere TLE were recruited from Comprehensive Epilepsy Program at the University of Florida. Patients were administered a comprehensive neuropsychological battery (including intellectual, language, memory, attentional, and executive function) during their inpatient stay. Patients underwent contemporaneous structural MRI scanning on a 3T Siemens Verio Scanner. Structural data included a T1-weighted magnetization prepared rapid acquisition gradient-echo (MPRAGE) sequence (1mm isotropic) for registration and T2-weighted turbo spin-echo (TSE) sequences (0.44 x 0.44 x 1.5mm) for hippocampal segmentation as part of clinical care. Hippocampal subfields, including CA2–CA3, DG, CA1, and the subiculum as well as the extrahippocampal structures of the entorhinal, perirhinal, and parahippocampal cortices, defined in the TSE images using the automatic segmentation of hippocampal subfields (ASHS) machine learning toolbox18 and a database of manual medial temporal lobe (MTL) segmentations. Results: The group comparison analysis suggested there was a statistically significant difference between groups (left TLE vs. right TLE) and hippocampal subfield volumes as determined by one-way ANOVA, only for the DG region (F(1,11) = [10.895], p = 0.007), with left TLE patients displaying reduced DG volume compared to right TLE patients. Correlations between the MRI volumetry and neuropsychological test scores were determined using Pearson's correlation coefficient analysis. In addition to a significant main effect of laterality, left hemisphere DG volume was positively correlated with performance on the WASI-II vocabulary subtest (r = .705, p = .007). Additionally, right CA2-3 volume was negatively associated with WASI-II FSIQ (r = -.594, p = 0.55). Importantly, right hemisphere MTL volumes were not associated with nonverbal memory performance. Conclusions: These results indicate that, partially consistent with the traditional view of material-specific information processing, patients with seizures originating the left temporal lobe exhibited reduced left-hippocampal volume (specifically in the dentate gyrus) and performed more poorly on a verbal vocabulary measure. The same relationships were not seen in the right hemisphere. While the lack of additional significant relationships between temporal lobe subfield volumes and cognitive measures may be an artifact of the relatively low sample size included in this preliminary data set, it may also serve to reaffirm the notion that language-based dominant hemisphere dysfunction is more readily apparent in TLE than is corresponding damage in the nondominant right hemisphere.

Poster

02/16/2024
03:30 pm - 04:45 pm
Room: Majestic Complex (Posters 61-120)

Poster Session 09: Epilepsy | Oncology | MS | Infectious Disease

Final Abstract #112

Dentate gyrus volume, but not other medial temporal lobe subregions, predict performance on select neuropsychological measures.

Sean Turner, University of Florida, GAINESVILLE, United States
Russell Bauer, University of Florida, Gainesville, United States

Category: Epilepsy/Seizures

Keyword 1: epilepsy / seizure disorders
Keyword 2: neuropsychological assessment
Keyword 3: hippocampus

Objective:

Neuropsychological (NP) assessment of surgical epilepsy patients plays a crucial role in localizing areas of dysfunction and predicting postoperative cognitive change. While the specialization of the dominant temporal lobe for language and verbal memory has long been established, considerable controversy exists about the memory functions of the language nondominant temporal lobe, since evidence of material-specific nonverbal memory impairment in right temporal lobe epilepsy (TLE) patients is inconsistent. One lingering question is whether NP test results would reflect with equivalent sensitivity the underlying pathophysiology responsible for language-dominant v. nondominant TLE. Here we sought to investigate whether hippocampal and extra hippocampal subfield volume determined from clinical MRI scans predicted lateralized neuropsychological test performance in a population with temporal-lobe epilepsy.

Participants and Methods:

Five patients with left-hemisphere TLE and eight patients with right-hemisphere TLE were recruited from Comprehensive Epilepsy Program at the University of Florida. Patients were administered a comprehensive neuropsychological battery (including intellectual, language, memory, attentional, and executive function) during their inpatient stay. Patients underwent contemporaneous structural MRI scanning on a 3T Siemens Verio Scanner. Structural data included a T1-weighted magnetization prepared rapid acquisition gradient-echo (MPRAGE) sequence (1mm isotropic) for registration and T2-weighted turbo spin-echo (TSE) sequences (0.44 x 0.44 x 1.5mm) for hippocampal segmentation as part of clinical care. Hippocampal subfields, including CA2–CA3, DG, CA1, and the subiculum as well as the extrahippocampal structures of the entorhinal, perirhinal, and parahippocampal cortices, defined in the TSE images using the automatic segmentation of hippocampal subfields (ASHS) machine learning toolbox18 and a database of manual medial temporal lobe (MTL) segmentations.

Results:

The group comparison analysis suggested there was a statistically significant difference between groups (left TLE vs. right TLE) and hippocampal subfield volumes as determined by one-way ANOVA, only for the DG region (F(1,11) = [10.895], p = 0.007), with left TLE patients displaying reduced DG volume compared to right TLE patients. Correlations between the MRI volumetry and neuropsychological test scores were determined using Pearson's correlation coefficient analysis. In addition to a significant main effect of laterality, left hemisphere DG volume was positively correlated with performance on the WASI-II vocabulary subtest (r = .705, p = .007). Additionally, right CA2-3 volume was negatively associated with WASI-II FSIQ (r = -.594, p = 0.55). Importantly, right hemisphere MTL volumes were not associated with nonverbal memory performance.

Conclusions:

These results indicate that, partially consistent with the traditional view of material-specific information processing, patients with seizures originating the left temporal lobe exhibited reduced left-hippocampal volume (specifically in the dentate gyrus) and performed more poorly on a verbal vocabulary measure. The same relationships were not seen in the right hemisphere. While the lack of additional significant relationships between temporal lobe subfield volumes and cognitive measures may be an artifact of the relatively low sample size included in this preliminary data set, it may also serve to reaffirm the notion that language-based dominant hemisphere dysfunction is more readily apparent in TLE than is corresponding damage in the nondominant right hemisphere.