Abstracts

Rationale: An estimated 200,000 patients are eligible for epilepsy surgery in the United States, yet less than 5000 procedures are performed annually, possibly due to reluctance to consent to open brain surgery. A minimally invasive stereotactic procedure may allow thermal ablation of seizure foci with less damage to surrounding tissue than resective surgery. We present results from four patients with intractable localization-related epilepsy that underwent MRI-guided stereotactic laser ablation (SLA) of their seizure foci. Methods: Four patients (10 y/o M with tuberous sclerosis (TS), 16 y/o F with mesial temporal sclerosis (MTS), 9 y/o M with hypothalamic hamartoma (HH), and 5 y/o M with left frontal cortical dysplasia (CD)) were consented for SLA. Under general anesthesia, patients were placed in a stereotactic head frame for creation of a cranial access via a 3.2mm twist drill hole. An MR-compatible laser applicator (1.6mm diameter) housing a 1cm diffusing tip optical fiber was placed into MRI visible target lesions and secured with a plastic bone anchor. Patients were then placed into a 1.5T MRI for real-time magnetic resonance thermal imaging (MRTI) during the ablation process. The FDA-cleared Visualase thermal ablation system, which includes a 15W 980nm diode laser and MRTI software were used to deliver an initial laser test dose (3-4W for 15-45 seconds) to confirm proper applicator position prior to thermal ablation doses (10W-12W for 45-120 seconds). MRTI was accomplished using PRF phase difference imaging techniques using a fast field echo (FFE, 256x128 acquisition matrix, TE=20ms, TR=38ms, flipangle = 30 deg) sequence. Safety limits (50 degrees C), which if exceeded, automatically terminated laser delivery, were placed near the margin of the desired thermal ablation zone to protect critical structures (optic tract, cerebral arteries, and hypothalamus). Post ablation T1-weighted plus gadolinium contrast images were acquired after completion of ablation cycles to confirm areas of thermal ablation. Results: SLA was successful in all cases without surgical complications. Post-contrast MRI exhibited a non-enhancing central coagulated region surrounded by a rim of enhancement typical of acute thermal injury that encompassed the targeted foci. The TS, MTS, HH and CD patients are all seizure free at 10 mo, 10 mo, 3 mo, and 2 weeks, respectively. Conclusions: MR-guided SLA offers advantages over conventional resective surgical approaches for removal of epileptic foci. The small diameter laser applicator provides a minimally invasive approach to reaching target foci including deep seated lesions. Additionally, use of real-time MRTI feedback control and estimates of thermal damage, allow excellent visualization of target and protection of critical structures during ablation process. Short term follow-up outcomes after SLA are encouraging and justify further studies.