Preoperative diagnosis of mesial temporal lobe epilepsy subtypes with in vivo MRI

  • Author / Creator
    Steve, Trevor A
  • Approximately 50 million people worldwide suffer from epilepsy, a chronic neurological disorder characterized by recurrent seizures. While most patients with epilepsy will respond to anticonvulsant medications, approximately one third are found to have drug resistant epilepsy. Mesial temporal lobe epilepsy (MTLE) is the most common clinical syndrome associated with intractable seizures.

    Pathological studies suggest that MTLE consists of multiple disease subtypes which have distinct clinical features and prognoses for surgical cure. However, MTLE subtypes currently can only be diagnosed postoperatively following surgical resection of the hippocampus. This method of characterizing MTLE subtypes has several limitations including incomplete evaluation of the symptomatic hippocampus, inability to assess the contralateral hippocampus, and analysis of subtypes at only one time point (the time of surgical resection). Therefore, our current understanding of MTLE subtypes remains incomplete. In vivo MRI holds the potential to improve our knowledge of MTLE subtypes by enabling preoperative diagnosis of these disease variants.

    The human hippocampus consists of multiple discrete regions (termed 'subfields') which have different histological features, synaptic connectivities, and putative functions. Several lines of evidence (including animal models and human neuropathology series) support the hypothesis that MTLE is a subfield-specific neurological disorder. While qualitative neuropathological studies suggest that MTLE is characterized by selective involvement of certain hippocampal subfields, it remains unclear whether quantifiable differences exist. We therefore conducted a systematic review and meta-analysis of the current literature in order to quantify subfield involvement in patients with MTLE. We found the greatest magnitude of neuronal loss in CA1 and relative sparing of the CA2 subfield, as suggested by the qualitative neuropathological literature.
    MTLE subtypes are currently diagnosed on the basis of neuronal density measurements from specific hippocampal subfields (CA1 and CA4). Despite advances in MRI spatial resolution, neuronal densities cannot be directly measured with this modality. In vivo MRI studies therefore use hippocampal subfield volumes (or areas) as a surrogate marker for subregional densities. We therefore examined the relationship between neuronal densities and subfield areas in our cohort of cadaveric hippocampi. Interestingly, our analysis did not reveal any correlation between neuronal densities and histology-defined subfield areas in control subjects. While our results suggest a complex relationship between these two measurements in normal hippocampi, several studies have demonstrated a positive correlation between neuronal densities and subfield areas in surgical specimens resected from patients with MTLE.

    Recent work has demonstrated the feasibility of mapping histologically defined hippocampal subfields onto ex vivo MR Images, but this has only been described in a single subject. In this thesis, we sought to determine if histological measurements from multiple cadaveric hippocampi could be used to develop a histologically validated segmentation protocol for the hippocampal body. We hypothesized that subfield transitions occur at predictable locations as a proportion of an anatomical structure (the stratum lacunosum moleculare, SLM) which is readily visualized with in vivo MRI. We found that the CA1/CA2 and CA2/CA3 subfield transitions could be accurately localized by using the mean proportion of the SLM for these boundaries as measured from histology. These data were used to obtain ex vivo MRI measurements of subfield areas (CA1, CA2, and CA3/CA4/DG) which were strongly correlated with the 'ground truth' histological areas of these same subregions.

    Our method is based on anatomical landmarks which are readily visible with in vivo MRI, which allowed us to directly translate this methodology to preoperative MR Images from patients with MTLE. We found that our method provided reliable hippocampal subfield measurements in healthy controls and subjects with MTLE. Hippocampal subfield volume measurements were then used to determine MTLE subtypes, which were consistent with the gold standard neuropathological subtype diagnoses in three out of five cases. Furthermore, analysis of subfield areas revealed significant variability of MTLE subtypes along the hippocampal long axis in several patients, which is consistent with previous autopsy studies. These results collectively suggest that our novel method can be reliably applied to determine disease subtypes in subjects with MTLE.
    In summary, our results support the hypothesis that MTLE subtypes can be accurately characterized with preoperative in vivo MRI. Further studies are required to explore the clinical correlates of these findings in a larger cohort of subjects and to evaluate the prognostic significance of these findings for patients with MTLE.

  • Subjects / Keywords
  • Graduation date
    Fall 2018
  • Type of Item
  • Degree
    Doctor of Philosophy
  • DOI
  • License
    Permission is hereby granted to the University of Alberta Libraries to reproduce single copies of this thesis and to lend or sell such copies for private, scholarly or scientific research purposes only. Where the thesis is converted to, or otherwise made available in digital form, the University of Alberta will advise potential users of the thesis of these terms. The author reserves all other publication and other rights in association with the copyright in the thesis and, except as herein before provided, neither the thesis nor any substantial portion thereof may be printed or otherwise reproduced in any material form whatsoever without the author's prior written permission.