Abstracts

Rationale: Stereotactic laser amygdalohippocampotomy (SLAH) is a novel highly selective ablative neurosurgical procedure for mesial temporal lobe (TL) epilepsy. We employed voxel-based lesion mapping (VBLM) analysis to determine which TL structures were affected by SLAH as compared with traditional open anterior TL resection procedures. Methods: VBLM analysis was conducted for multiple TL structures using post-surgical 3T MRI scans (T1 MPRAGE) obtained on a Siemens Trio for the first 26 patients undergoing SLAH and 3 consecutive open resection cases at Emory University. Ablation/resection (i.e., lesion) size percentages were created for each region of interest (ROI) by tracing the lesion from the post-surgical MRI scan and determining the proportional size of the lesion as compared to the defined ROI from a standardized brain atlas. We present the percentage of patients with volumetric changes for each of these assessed TL structures, exploring the variability within the SLAH procedure and also comparing the extent to which open and SLAH procedures differ with regards to their damage zones. Results: All SLAH patients exhibited volumetric decline in their amygdala and hippocampus. However, half of the subjects also experienced volumetric decline in their entorhinal cortex (13/26) and the majority also exhibited at least mild decline in parahippocampal gyrus volumes (20/26). Of note, none of the SLAH patients experienced any change in their temporal pole, fusiform gyrus, or their inferior, middle, or superior temporal gyri, although the open resection patients also experienced decline in all but the superior temporal gyrus. Even here, the open resection patients did show alteration of the anterior portion of the superior temporal gyrus, which is included in the temporal pole. Conclusions: While it has been argued that better cognitive outcomes with SLAH likely results from sparing broader TL regions ipsilateral to seizure onset as compared to open resection, this is the first study to objectively document the differences between these procedures with VBLM analyses. Essentially, the SLAH procedure in all cases was restricted to medial TL structures and did considerably less damage even in these structures that fell beyond the amygdala and hippocampus. Additionally, it is also clear that the extent and location of the damage zone can substantially vary across SLAH patients, and it is possible that such variation could affect both cognitive and seizure freedom outcome, and should be the focus of future research.  Funding: The current project was carried out with grant support received by Dr. Drane from the NIH/NINDS (R01NS088748, K02NS070960) and Medtronic, Inc. (A1225797BFN).