Authors :
Presenting Author: Raunak Singh, MD – Mayo clinic, Rochester, Minnesota
Shelja Sharma, MD – Mayo clinic, Rochester, Minnesota
Luis Alcala-Zermeno, MD – Columbia University
Gamal Eldin Osman, MD – Mayo Clinic
Nicholas Gregg, MD – Mayo clinic, Rochester, Minnesota
Keith Starnes, MD – Mayo Clinic, Rochester MN, USA.
Kai Miller, MD, PhD – Mayo Clinic
Jamie J Van Gompel, MD – Mayo Clinic, Rochester MN, USA.
Gregory Worrel, MD,PhD – Mayo Clinics
Brian Lundstrom, MD PhD – Mayo clinic, Rochester, Minnesota
Rationale:
Invasive neuromodulation (I-NM) is a non-curative therapy for drug-resistant epilepsy (DRE), but there is limited literature that compares long-term effectiveness between devices.
Methods:
This single-center retrospective study compared 2-lead deep brain stimulation (DBS-ANT, DBS-CM), 4-lead DBS (DBS-ANT/CM, DBS-ANT/Pulvinar, DBS-Cortical/Subcortical), responsive neurostimulation (RNS) and vagus nerve stimulation (VNS) used to treat DRE. Patients with new devices implanted between January 2021 to December 2024 at Mayo Clinic (Rochester, MN), with at least 6 months follow-up were included. 6 patients were excluded due to ongoing clinical trials: 4 (NAUTILUS) RNS and 2 (START) DBS-RC+S.
Results:
176 patients (75 women, 43%) (50 children, 28%) were included. Patients receiving 2-lead DBS and RNS were older than 4-lead DBS (p=0.005 and p=0.04, respectively) and VNS (p=0.0007 and p=0.01, respectively) recipients. 41% of patients (n=83) with DBS or RNS implants had a previous history of VNS. 23% of patients (n=41) had previous epilepsy surgery. Baseline seizure frequency did not differ between 2-lead DBS (12/month), 4-lead DBS (30/month), RNS (11/month) or VNS (19/month) (p >0.05). Median seizure reduction (MSR) at 3, 6 and 12 months across all devices was 33%, 40% and 46%, respectively. MSR at those time points for 2-lead DBS was 50%, 50% and 58%; for 4-lead DBS 50%, 59% and 70%; for RNS 29%, 40% and 56%; and for VNS 25%, 20% and 35%, respectively. All groups achieve significant seizure reduction at all time points with respect to baseline (p< 0.05). 4-lead DBS showed significant improvement compared to VNS at 6 months (p=0.03) and 12 months (p=0.009). Responder rates (RR) at 3, 6 and 12 months were 43%, 46% and 57% compared to baseline (p< 0.0001). At those time points, RR for 2-lead DBS was 54%, 51% and 65%; for 4-lead DBS 52%, 63% and 70%; for RNS 39%, 44% and 61%; for VNS 36%, 40% and 50%, respectively. Responders did not differ significantly between different device groups at any time point (p >0.05). Seizure frequency was not different between patients with generalized and mixed epilepsies (n=60) compared to focal and multifocal epilepsies (n=86) (p >0.05).
Conclusions:
I-NM decreases seizure burden for DRE with respect to baseline. Responder rates improve with time for all device groups. 4-lead DBS performed better than VNS at 6 months and 12 months while no significant differences were observed between other devices at different time points. Further, data may be needed to identify differences between devices for specific epilepsy types, seizure types and comorbidities.
Funding:
NINDS R01NS129622