What defines response to vagus nerve stimulation in children with drug-resistant epilepsy? A prospective cohort study from the CONNECTiVOS collaboration

Abstract

Objectives: Responsiveness to vagus nerve stimulation (VNS) in children with drug-resistant epilepsy (DRE) is often defined based on reduction in seizure frequency, typically at the 50% threshold, with limited consideration to the effects of therapy on seizure severity and health-related quality of life (HRQoL). In the current report, we sought to better characterize the effects of VNS beyond seizure frequency in children with DRE. Methods: Sixty-seven children from the Connectomic profiling and Vagus nerve stimulation Outcomes Study (CONNECTiVOS) database, a multicenter study including children aged 0-18 who underwent VNS at eight North American centers, were included. Data were collected prospectively at baseline and 6, 12, and 24 months after VNS. Seizure outcomes were assessed using the 50% threshold, percentage change in seizure count and change on a frequency timescale, a categorical measure of seizure occurrence (e.g., daily to weekly seizures). The Seizure Severity Questionnaire and the Quality of Life in Childhood Epilepsy were also collected. Linear mixed models were constructed to study longitudinal changes in seizure severity and HRQoL. Results: Among 67 children, 55.2% experienced >50% reduction in seizures, whereas only 37.3% demonstrated reductions in seizure timescales. Notably, 17.9% experienced no reduction in timescales despite meeting the 50% threshold. Furthermore, 31.9% of children experienced a meaningful improvement in seizure severity without any reduction in seizure timescales. HRQoL improvements were driven by the reduction in timescales (mean yearly increase of 3.65; 95% confidence interval [CI]: 0.71-6.52, p = 0.023), rather than responsiveness based on the 50% threshold or percentage change in seizure counts (p > 0.05). Reduced overall seizure severity was also independently associated with higher HRQoL after VNS (p = 0.035). Significance: The conventional 50% responder threshold failed to capture meaningful gains in HRQoL, which aligned more closely with seizure timescales. Furthermore, nearly one-third of children realized improvements in seizure severity without any reduction in timescales. Improvements in both seizure frequency timescales and seizure severity drive postoperative HRQoL gains.