Abstracts

Rationale: Seizures occur in 20% to 45% of patients with brain arteriovenous malformations (AVM), and are the most common presenting symptom in unruptured AVMs. Modern treatment modalities for AVMs include micro surgery (MS), stereotactic radiosurgery (SRS), and endovascular embolization (EVE), all three of which are effective in achieving hemodynamic control of the AVM. However, the effectiveness of each modality in controlling AVM-related epilepsy remains a topic of study. No previous studies have provided detailed, long-term seizure outcomes (Engel scale) after multimodal AVM treatment with SRS+EVE. We retrospectively analyzed a cohort of 111 AVMs treated at our institution with either SRS or SRS+EVE. Our aims were to evaluate the impact of SRS+EVE compared to SRS on long-term seizure control and compare the rates of antiepileptic drug (AED) use and de novo seizures between treatment groups.

Methods: Retrospective review of a database containing 126 patients who underwent Gamma Knife stereotactic radiosurgery for the treatment of an AVM at the University at Buffalo from 1999 to 2019. Patients were excluded from the analysis if there was a lack of detailed seizure outcomes after AVM treatment (n = 15). Seizure outcome was determined using a modified Engel Epilepsy Surgery Outcome Scale consisting of three groups: (1) absence of all seizures after treatment for a duration of at least 1 year (Engel Class IA), (2) absence of disabling seizures after treatment for a duration of at least 1 year (Engel Class IB), or (3) outcomes other than Engel IA or IB (Engel Classes II-IV).

Results: The median patient age was 43 years (range 0-90). The mean follow-up duration was 73.2 months (6.1 years) and the median was 70.0 months (5.9 years), with a range of 4 to 220 months. The primary presenting symptoms were isolated headache in (30.6%), and seizure in 17 patients (15.3%). Seizure outcomes: Twenty-nine of 38 (76.3%) patients had their AVM successfully treated with SRS without ever experiencing a seizure, compared to 50/73 (68.5%) in patients who received SRS+EVE (p = 0.508). In patients who initially presented with seizure (n = 17), seizure freedom (Engel Class IA) was achieved in 1/5 (20.0%) patients with SRS, vs. 5/12 (41.7%) with SRS+EVE (p=0.600). Seizure reduction (Engel Class IA and IB) was seen in 3/5 (60%) patients who received SRS and 11/12 (91.6%) patients who received SRS+EVE (p=0.191). AED use: There was no statistically significant difference in the use of AEDs at the last follow-up between treatment groups (p=0.210). Of the entire cohort, 71 patients (64.5%) were not taking any form of AED at the last follow-up. New-onset seizures: Seventeen patients experienced new-onset seizures after AVM treatment. Seizures began after SRS in 5 patients, after EVE in 3 patients, and after SRS+EVE in 9 patients. At last, follow-up, 4 patients achieved seizure freedom (Engel Class IA), 8 patients suffered only non-disabling seizures (Engel Class IB), and 5 suffered from some disabling seizures (Engle Class II-IV).

Conclusions: Our preliminary conclusion showed that there is no difference with regards to seizure control between AVM patients treated with SRS alone vs SRS and embolization.

Funding: Not applicable