Multi-voxel analysis of hippocampus

Author(s)

Bruno Augusto Goulart Campos

• Document type: Doctoral Thesis
• Press: Campinas, SP.
• Institution: Universidade Estadual de Campinas (UNICAMP). Faculdade de Ciências Médicas
• Defense date: 2016-04-29

Examining board members

Fernando Cendes; Clarissa Lin Yasuda; André Luiz Ferreira Costa; Luiz Eduardo Betting; Veriano Alexandre Junior

• Advisor: Fernando Cendes

Abstract

Objectives: Use the multi-voxel spectroscopy MRI to study the biochemical markers (NAA, Creatine and Hill) in patients with Temporal Lobe Epilepsy with frequent and infrequent attacks, segmenting the hippocampus in subareas (anterior, middle and posterior). Methods: 15 patients with Temporal Lobe Epilepsy with infrequent seizures and 14 patients with frequent seizures (more than 3 per year) were studied, regularly followed in a tertiary epilepsy centre (Epilepsy Clinic of Campinas State University). These were compared with 23 healthy subjects and all were subjected to study imaging and spectroscopy by MRI. The NAA values were quantified, Creatine, Choline and the ratio NAA / creatine was also calculated in the three segments of the hippocampus (anterior, middle and posterior). Results: NAA metabolites, Choline and Creatine were tested by analysis of variance (ANOVA) between the groups, showing a significant reduction of NAA in the three segments (anterior, middle and posterior) of the ipsilateral hippocampus to the EEG (p = 0.001, p = 0.01 and p = 0.001 respectively) and the middle and posterior segments contralateral to EEG (p = 0.01 and p = 0.02), but not in the contralateral anterior segment (p = 0.069). Analyzing the ratio NAA / Creatine, a significant reduction was observed only in the middle and posterior segments of the ipsilateral hippocampus (p = 0.04 and p = 0.01) but not in the anterior segment (p = 0.11) or all contralateral segments (p = 0.81, p = 0.11 and p = 0.15). Discussion: The study showed that there is a significant reduction of NAA in all segments of the hippocampus ipsilateral to the EEG changes, but in the contralateral hippocampus it occurs only in the medium and posterior segment. Still, the ratio NAA / creatine was significantly reduced only in the middle segment and posterior ipsilateral to the change of EEG. Although there seems to be a greater reduction in subsequent segments based on the ratio NAA / creatine and contralateral side, this study has a high frequency of patients with hippocampal atrophy and long duration of epilepsy, which could disturb the analysis of metabolites. Conclusions: We note a greater reduction of NAA in the middle and posterior segments of the hippocampus. This may indicate that these areas are more important in the genesis and maintence of the temporal lobe epilepsy. In addition, this may suggest positioning the single-voxel spectroscopy to the most posterior segment (AU)