Abstracts

Rationale: Corpus callosotomy is an established therapy for drug-resistant epilepsy complicated by atonic seizures. Meta-analyses show open callosotomy is associated with a 70% rate of drop attacks cessation. Potential complications of traditional open callosotomy may be due to the surgical approach, rather than disconnection itself. MR-guided laser interstitial thermal therapy (MRgLITT) is a minimally invasive alternative to traditional open disconnection, however a lower extent of disconnection with MRgLITT is a perceived drawback of this approach. Structural MR imaging before and after MRgLITT callosotomy is crucial to understand the extent of disconnection associated with this procedure and how it relates to outcomes.

Methods: We identified five pediatric patients who underwent MRgLITT callosotomy with pre and post-operative diffusion imaging, ages 5 to 21 (median age, 16). We computed probabilistic tractography from high-definition fiber tractography (HDFT) sequences before and after (3-month post-op) corpus callosotomy for each of the major callosal white matter pathways, including the forceps minor, forceps major, tapetum, and body. Differential tractography was used to quantify the degree of disconnection following callosotomy using a threshold of 30% difference in anisotropy. Fractional anisotropy and tract volume was compared between baseline and follow-up. Retrospective electronic medical record data collection was used to collect clinical variables. 

Results: All patients had a diagnosis of Lennox-Gastaut syndrome (LGS) and suffered drug-resistant atonic seizures prior to MRgLITT corpus callosotomy. Three patients underwent complete callosotomy, and two patients underwent completion of prior open microsurgical anterior two-thirds callosotomy. Patients did not suffer any acute complications following callosotomy. Four patients had at least 12-months of follow-up and average length of follow-up was 13 months. A total of 80% (n=4/5) of patients were free from atonic seizures at follow-up, while the remaining 20% had experienced only an improvement in atonic seizure frequency at 3-month follow-up. We show that callosal fibers can be ablated with 3 to 4 laser trajectories for complete callosotomy, and with 1 to 2 for completion of a partial callosotomy. Imaging acquired at 3 months post-op shows complete callosotomy patients had 71% decrease in baseline interhemispheric callosal white matter on average. Diffusion anisotropy was also reduced, suggesting largescale microstructural degradation following axonal injury.

Conclusions: These results support existing evidence for MRgLITT as a safe and effective minimally invasive surgical option for interhemispheric disconnection of the corpus callosum in select pediatric patients experiencing drug-resistant atonic seizures (complete corpus callosotomy or completion of prior open callosotomy). The use of probabilistic tractography is a useful imaging modality to quantify the degree of disconnection when anatomical imaging is inconclusive. Here, we use tractography to quantify large-scale changes in volume and fractional anisotropy of interhemispheric corpus callosum axonal pathways 3-months after LITT callosotomy.

Funding: Not applicable