Deep Brain,Deep Impact: Rethinking Pediatric Epilepsy Treatment With Deep Brain Stimulation of the Centromedian Nucleus

Abstract

Original Article Citation

Mithani K, Niazi F, Suresh H, Alrumayyan Y, Rayco ER, Ochi A, Otsubo H, Kerr E, Breitbart S, LeBlanc-Millar A, Gadgil N, Raskin JS, Weil AG, Hadjinicolaou A, Iorio-Morin C, Weiss S, Jain P, Sham L, Donner E, Fasano A, Gorodetsky C, Ibrahim GM. Deep brain stimulation of the centromedian nucleus for drug-resistant epilepsy in children: Quality-of-life and functional outcomes from the CHILD-DBS registry. Epilepsia. 2025 Apr 1. doi: 10.1111/epi.18393. Epub ahead of print. PMID: 40167366.

Objective: Deep brain stimulation of the centromedian nucleus of the thalamus (CM-DBS) is an investigational, off-label treatment for drug-resistant epilepsy (DRE) in children. Although emerging evidence supports its safety and efficacy for select indications, the effect of CM-DBS on quality of life and functional outcomes such as school attendance has not been studied. Here, we analyzed data from the prospective CHILD-DBS (Child and Youth Comprehensive Longitudinal Database for Deep Brain Stimulation) to examine the impact of CM-DBS on patient- and caregiver-reported outcomes. Methods: Twenty-two children and youth underwent bilateral CM-DBS. Caregiver–child dyads completed surveys related to seizure frequency, seizure severity, quality of life, and school attendance at baseline, 6 months, and 1 year postsurgery. Simulated volumes of tissue activation were analyzed to identify optimal stimulation targets associated with treatment outcome. Results: Of 22 children, 10 experienced ≥50% reduction in seizure frequency (mean reduction = 66.7 ± 17.3%), one exhibited a modest benefit (37.5% reduction), and the remaining 11 experienced no change. The majority (73% of patients) exhibited a clinically important reduction in seizure severity, including six children who did not demonstrate any change in seizure frequency. Only those who experienced a reduction in seizure frequency demonstrated significant improvements in general health and overall quality of life. Furthermore, we observed an increase in school attendance across participants 1 year after CM-DBS. Significance: CM-DBS can lead to reduction in seizure burden concurrent with improvements in quality of life and relevant functional outcomes in children with DRE. These findings further our understanding of the impact of CM-DBS on meaningful outcomes for children and caregivers.

Commentary

Intractable epilepsy, failure to respond to two or more antiseizure medications, affects approximately one-quarter of children with epilepsy. ¹ Neuromodulation using deep brain stimulation (DBS) is an off-label, investigative therapy in the United States for the palliative treatment of drug-resistant epilepsy in children who are not otherwise candidates for a definitive or resective neurosurgical intervention. While studies have evaluated the potential benefits of epilepsy surgery on health-related quality of life (HRQOL) in children, ² the benefit of neuromodulation with DBS has not been fully elucidated. There is evidence in adults for stimulation of the anterior nucleus of the thalamus ³ and centromedian nucleus (CM), ⁴ but available pediatric data is less robust. The ESTEL trial of CM-DBS in adults with Lennox-Gastaut Syndrome (LGS) demonstrated an approximately 50% responder rate with seizure reduction of 50%. ⁴ Pediatric retrospective series of CM-DBS have shown response rates of up to 83% experiencing >50% reduction in seizure frequency. ⁵

The Child and Youth Comprehensive Longitudinal Database for DBS (CHILD-DBS) is a prospective, multicenter registry of six pediatric epilepsy centers in Canada and the United States performing DBS implantation. ⁶ The authors report on HRQOL and functional outcomes, focusing on seizure frequency, seizure severity, and school attendance. One goal of this article was to further clarify benefits of CM-DBS on daily functioning and QOL in children using the largest prospective pediatric DBS cohort to date.

All children underwent bilateral implantation with either Medtronic 3387 or 33015 SenSight Directional leads and all were implanted with Medtronic Percept PC implantable pulse generators with surgical decision making per individual center standard of care. Seizure outcome questionnaires were collected at baseline, 6 months, and 12 months postoperatively, obtaining frequency of the most debilitating seizures over the preceding month, and considered “responders” if they had a reduction in seizure frequency at last follow-up. Caregivers completed the 24-item Seizure Severity Questionnaire (SSQ) at each time point. Quality of life and functional outcomes were assessed using the Quality of Life in Childhood Epilepsy (QOLCE) questionnaire, a 91-item questionnaire that evaluates 5 subdomains: physical, cognitive, and social function, emotional well-being, and behavior. Additionally, caregivers were asked to report the number of days of school missed in the month prior to each time point. Finally, the authors performed “sweet spot” mapping—probabilistic stimulation mapping to identify areas within the CM associated with greatest seizure reduction based on reconstruction of each participant's DBS implantation and volume of tissue activation (VTA) using computer-based simulation.

Participants included 22 children and adolescents (mean age 13.4 years, range 5-18 years) with long-standing and refractory epilepsy, with mean duration of epilepsy prior to surgery of 11.4 years and an average epilepsy onset at age 2 years (SD 2.5 years). This cohort included children with comorbid autism spectrum disorder (n = 7, 32%) and attention deficit hyperactivity disorder (n = 2, 9%), and over half of the participants (59%) had a diagnosis of a developmental and epileptic encephalopathy (DEE), including LGS (8), infantile epileptic spasms syndrome (2) and early infantile DEE (2). Etiologies were varied including genetic or syndromic disorders, perinatal and structural etiologies, and unknown etiology. It is notable that 18 of 22 (82%) had prior VNS implantation.

A significant reduction in seizures was seen over time with a bimodal distribution, with 10 (45%) experiencing ≥50% reduction (mean reduction 66.8 ± 17.3%), one a reduction of 37.5%, and 11 experiencing no change. There was no association between seizure reduction at 1 year follow-up and any other factors including age at implantation, age at diagnosis, underlying DEE, or a specific DEE/etiology. The presence of generalized paroxysmal fast activity (GPFA) on preoperative EEG was significantly associated with a response to CM-DBS (chi-squared p = .034). Interestingly, among 15 children-caregiver pairs who completed the SSQ at 1 year, there was a significant reduction in overall seizure severity seen in 73% (n = 11), while of those, 6 (55%) had no reduction in seizure frequency.

“Sweet spot” mapping identified bilaterally paired spots in the anterolateral CM associated with greater reduction in seizure frequency at 1-year follow-up in addition to a posteromedial sweet spot in the right CM. The authors found that overlap between VTAs and these sweet spots explained 77% of the variance in percent change in seizure frequency at 1 year follow-up (R² = 0.77, p < .001).

Expectedly, those children who had a reduction in seizure frequency had a significant improvement in overall QOL, seen as a total QOLCE score change between baseline and 1-year follow-up (p = .031), compared to those who did not experience seizure reduction (p = .58). Additionally, caregivers reported a significant reduction in number of school days missed at 1-year postoperatively compared to baseline in responders compared to nonresponders, with an average of 4.2 fewer missed school days (Wilcoxon p = .047).

The responder rate was approximately 45%, which is in keeping with previously reported prospective studies of CM-DBS in adults with refractory epilepsy, ⁴ and similar to findings from the ADVANCE trial, a partially randomized patient preference trial of add-on DBS versus continued VNS. ⁷ The authors noted that the response to DBS was lower than what had previously been found in systematic reviews of retrospective series, ⁵ but suggest that their findings likely more accurately reflect real-world data based on their prospective data collection and use of standardized measures.

One limitation of this cohort is the severity and heterogeneity of their epilepsy, which could potentially explain the lower response rate seen in the CHILD-DBS study compared to the ESTEL trial, where the participants all carried a diagnosis of LGS. ⁴ Certainly, as GPFA was a marker of likelihood to respond, participants with LGS might have a better response to CM-DBS; however, it is possible that this study was underpowered to identify an effect of etiology on response. Future studies could further clarify any differences in response based on the underlying DEE or etiology.

Another potential area to explore would include effects of CM-DBS on sleep and mood and behavior in children with refractory epilepsy given the potential impact on QOL that these domains have. As comorbid depression and anxiety are common in children with epilepsy, ⁸ effects of DBS on mood in children would be worth characterizing given that there exists the potential concern for depression as a side effect of DBS, ³ although, this has not been fully proven ⁹ and DBS can be used for treatment refractory depression. ¹⁰

Improved seizure frequency is associated with better quality of life. DBS remains off-label in pediatrics, limiting DBS utilization until children and adolescents have been living with debilitating, refractory seizures for an extended period of time, and only if they have access to a pediatric epilepsy center, rather than DBS existing as another therapeutic option for all who could benefit. The findings of the CHILD-DBS study are important to note improvement in QOL, including school attendance seen with seizure reduction. A significant proportion had a reduction in seizure severity which can also be meaningful, even in the absence of reduced seizure frequency, particularly if injurious seizures are fewer. Earlier introduction of neuromodulation may optimize seizure control as well as quality of life in our patients.