The impact of vagus nerve stimulation on the most disabling seizures: A retrospective study in adults with drug-resistant epilepsy

Comprehensive Summary

The study aims to visualize the long-term efficacy of vagus nerve stimulation in responders (R-VNS) and non-responders (NR-VNS) in adults with drug-resistant epilepsy. Previous research suggested that responders and non-responders differ in underlying brain network organization, but most studies used EEG or fMRI and focused on limited frequency ranges. To address this, the researchers analyzed preoperative resting-state magnetoencephalography functional connectivity, filtered into seven frequency bands from delta to fast ripple, and whole-brain functional connectivity using the Corrected Amplitude Envelope Correlation software. Using this database, NR-VNS patients showed widespread connectivity abnormalities compared with control and R-VNS patients, especially in the theta region, where they had 370 enhanced connections instead of 170. There was also a significant disconnectivity in NR-VNS patients across gamma, alpha, and delta frequencies. These indicate that R-VNS patients have a more normalized pattern of brain connectivity and network activity across different oscillations, such as alpha and gamma, prior to treatment.

Outcomes and Implications

This research has meaningful implications because it allows for more precise patient selection for VNS therapy. Visualizing that NR-VNS patients already exhibit widespread functional connectivity abnormalities before implantation suggests that VNS may be less effective in individuals whose brain networks are profoundly disordered. In contrast, R-VNS patients show connectivity patterns closer to those of healthy controls, particularly in the alpha and beta bands, suggesting a more stable and “treatable” brain network state. VNS is most effective for treating drug-resistant epilepsy when neural circuits retain structural integrity and already show synchronized activity. This allows preoperative MEG to serve as a biomarker to predict VNS responsiveness, thereby improving patient outcomes, reducing unnecessary interventions, and optimizing individualized care.