Abstracts

Rationale: Laser interstitial thermal therapy (LITT) is increasingly used for the ablation of the hippocampus and amygdala in mesial temporal lobe epilepsy (MTLE). Like traditional microsurgical resections, this technique also carries some risk for visual field (VF) deficits. Here we quantify the visual deficits resulting from LITT for MTLE in a cohort of 45 patients, with post-operative formal visual field assessments. Methods: The Humphrey field analyzer, using a 30-2 SITA-Standard protocol was used to assess visual function between 6 weeks and 6 months post-procedure. Only patients with post op full field (30-2) Humphrey visual fields (HVF) assessments, were analyzed. Masks of the ablation volume (enhancement margin on immediate post-ablation MRI) were co-registered into standardized space using FreeSurfer’s combined non-linear volumetric surface registration method. The percentage of significant pattern deviation test points per HVF quadrant was assigned to each white matter voxel included in that individual’s ablation mask. t-tests were used to create a statistical heatmap of all ablated voxels associated with HVF deficits. Additionally, binary maps of voxels that were included solely in the ablation masks that produced HVF deficits were created in the template space. These were then morphed back to individual patient spaces and used as seeds for DTI tractography. Results: 45 of these patients had postoperative HVFs. Of these, 16 were excluded from further analysis due to prior significant visual field deficits, a poor ability to participate in HVF test, or both. Of the remaining 29, 19 had left hemisphere (LH) LITT, and 10 had right hemisphere (RH) LITT. Only 1 patient developed a complete contralateral superior temporal quadrantanopsia. 31% of patients (n=9) developed a HVF deficit of more than 20% of the contralateral superior temporal quadrant. In 8 other patients, the deficit was substantially smaller. For the entire patient cohort, 17.7% (±30.0%) and 20.3% (± 29.0%) of test points in the contralateral superior temporal quadrant showed significant pattern deviations for OS and OD, respectively.  A volumetric analysis revealed that 55.6% (±20.6%) of the left and 56.4% (± 20.3%) of the right hippocampus was ablated. Grouped analysis of the ablation volumes revealed that ablations extending into the white matter superior and lateral to the hippocampus were more likely to results in VF deficits. An analysis linking ablations to pre-operative tractography, is ongoing. Conclusions: To the best of our knowledge, this is the largest such analysis of the impact of LITT on visual function to date. The incidence of major VF deficits following LITT is low (3%) but a higher fraction (31%) developed a partial VF deficit. These numbers are considerably lower than previously reported VF deficits after ATL or amygdalo-hippocampectomy (50-80%). Funding: Not applicable