Automated trajectory planning for laser interstitial thermal therapy in mesial temporal lobe epilepsy

Summary Objective Surgical resection of the mesial temporal structures brings seizure remission in 65% of individuals with drug‐resistant mesial temporal lobe epilepsy (MTLE). Laser interstitial thermal therapy (LiTT) is a novel therapy that may provide a minimally invasive means of ablating the mesial temporal structures with similar outcomes, while minimizing damage to the neocortex. Systematic trajectory planning helps ensure safety and optimal seizure freedom through adequate ablation of the amygdalohippocampal complex (AHC). Previous studies have highlighted the relationship between the residual unablated mesial hippocampal head and failure to achieve seizure freedom. We aim to implement computer‐assisted planning (CAP) to improve the ablation volume and safety of LiTT trajectories. Methods Twenty‐five patients who had previously undergone LiTT for MTLE were studied retrospectively. The EpiNav platform was used to automatically generate an optimal ablation trajectory, which was compared with the previous manually planned and implemented trajectory. Expected ablation volumes and safety profiles of each trajectory were modeled. The implemented laser trajectory and achieved ablation of mesial temporal lobe structures were quantified and correlated with seizure outcome. Results CAP automatically generated feasible trajectories with reduced overall risk metrics (P < .001) and intracerebral length (P = .007). There was a significant correlation between the actual and retrospective CAP‐anticipated ablation volumes, supporting a 15 mm diameter ablation zone model (P < .001). CAP trajectories would have provided significantly greater ablation of the amygdala (P = .0004) and AHC (P = .008), resulting in less residual unablated mesial hippocampal head (P = .001), and reduced ablation of the parahippocampal gyrus (P = .02). Significance Compared to manually planned trajectories CAP provides a better safety profile, with potentially improved seizure‐free outcome and reduced neuropsychological deficits, following LiTT for MTLE.