Efficacy and safety following two or more years of vagus nerve stimulation (VNS Therapy) in pediatric patients with drug‐resistant epilepsy enrolled in a Russian VNS Registry

Abstract Introduction Following approval in 2009 of vagus nerve stimulation (VNS Therapy) for drug‐resistant epilepsy (DRE) in the Russian Federation, this is the first multicenter study across Russia to evaluate the safety and efficacy of adjunctive VNS Therapy. Methods The retrospective, observational registry included 58 pediatric patients with DRE (5–17 years old at implantation) who had ≥2 years of VNS. To ensure a robust evaluation process, changes in seizure frequency were evaluated for all seizure types as well as “most disabling” seizures (defined as seizures accompanied by falls, physical trauma, and/or incontinence in the absence of preventative measures). Results With 2 years of VNS Therapy, 37 of 49 patients (76%) experiencing the most disabling epileptic seizures had a >50% decrease in frequency of such seizures, and 16 (33%) reported no longer experiencing the “most disabling” seizure type. In addition, based on the McHugh Outcome scale, VNS Therapy had a positive outcome on both frequency and severity of all epileptic seizure types, with a >50% decrease in frequency of all epileptic seizure types noted in 37 of 58 patients (64%), and 31% of patients had a Class I outcome, including 11 patients (19%) who achieved seizure freedom. VNS Therapy also had a positive effect on the frequency of status epilepticus: 13 patients (22%) had status epilepticus prior to implantation with a mean rate of 9.4 ± 17.7 events per year (range, 0–52), and after VNS Therapy, only one patient continued to experience status epilepticus (at 1 event per 4–6 months). VNS Therapy had an acceptable safety profile and no adverse events leading to VNS discontinuation were reported. Conclusions The results demonstrate that VNS Therapy is being safely and effectively applied to pediatric patients in the Russian healthcare system.


INTRODUCTION
Approximately 10.5 million children worldwide experience epileptic seizures (Guerrini, 2006;Jain & Arya, 2021;Moshe et al., 2015), and an estimated one third of the patients are resistant to antiseizure medication (ASM) resulting in high morbidity and mortality (Kwan et al., 2011).
Long-term recurrent seizures in children have a negative impact on their overall well-being (including their physical and cognitive development), sleep cycles, mood, and quality of life and place a heavy burden on the patients, their families, and caregivers (Austin & Caplan, 2007;Guerrini, 2006;Hansen et al., 2016;Jones et al., 2010).
Vagus Nerve Stimulation Therapy (VNS Therapy ® ) involves intermittent electrical stimulation of the left cervical vagus nerve leading to stimulation of the "vagus afferent network" (Hachem et al., 2018). To date, clinical studies conducted in various countries have demonstrated that following 24 months of adjunctive VNS Therapy, over 40% of pediatric patients with DRE experience ≥50% reduction in baseline seizure frequency and the response rate increases with duration of VNS Therapy (Orosz et al., 2014). Studies have also demonstrated that VNS Therapy in pediatric patients is associated with a reduction in seizure severity and seizure duration, as well as improvements in mood and certain domains of cognition (Orosz et al., 2014).
About 25 centers in Russia are currently providing VNS implantation procedures and more than 40 centers provide VNS system programing. However, the number of patients with access to VNS treatment is limited in Russia. This is due to insufficient funding of the Russian healthcare system that has led to a large gap in the provision of medical care, limited availability of modern diagnostic equipment, limited use of advanced medical technologies, an average salary for healthcare workers that is a quarter lower than the average for the national economy, and unfortunately, an overall lower life expectancy of the general Russian population compared with other industrialized nations (Ivanov & Suvorov, 2021 have medical records with information on seizure types, frequency, and severity for at least 2 months prior to implantation; achieve an output current of at least 1.5 mA within 9 months of implantation; and have medical records with data from follow-up assessments for at least the first 2 years of VNS treatment. A target of ≥1.5 mA output current was selected as it is considered to be the minimally effective current for positive outcomes (LivaNova, 2021). At an under-therapeutic output current, the probability is lower to gain seizure freedom (Fahoum et al., 2022).
Information on follow-up assessments included seizure types, seizure frequency, and seizure severity; adverse events and serious adverse events; use of other ASMs; changes in VNS Therapy parameter settings; and use and effectiveness of the external magnet.
Patients who underwent epilepsy surgery in the first year after VNS implantation or participated in another clinical trial within the first 2 years of implantation were ineligible to enroll in the registry.

Assessment of "most disabling" types of seizures and "all" types of seizures
Seizure frequency was calculated based on patient and caretaker reports. To ensure a robust evaluation process, changes in seizure fre-quency were evaluated for "all" seizure types as well as "most disabling" seizures.
The frequency dynamics of "all" seizure types were assessed using the McHugh Outcome scale that is a modification of the Engel Epilepsy Surgery Outcome Scale designed to assess outcomes after epilepsy surgery (McHugh et al., 2007). The McHugh is a scale of choice of VNS Therapy experts as it has been tailored to evaluate the outcomes following VNS Therapy to not only assess reduction in seizure frequency but also incorporate improvement in ictal and postictal seizure severity and benefits from use of the external magnet-both of which are very important VNS outcome measures (McHugh et al., 2007).
The "most disabling" seizures were defined as seizures accompanied by falls, physical trauma, and/or incontinence, in the absence of preventative measures.

Seizure severity assessment
To assess the effects on seizure severity, the validated National Hospital Seizure Severity Scale (NHS3) was adapted for this study (O'Donoghue et al., 1996).

Statistical analysis
Summary descriptive statistics were generated for continuous variables, and frequencies and percentages were generated for categorical variables. There was no imputation of missing data.
Pairwise comparison of means was performed using analysis of variance (ANOVA). Any significant difference was determined using Tukey's honest significant difference (HSD) test for multiple comparisons.

Demographics and baseline characteristics
The registry included 58 patients with 31 (53%) females and 27 (47%) males (Table 1). The mean age at implantation was 11 ± 3 years (range, 5-17 years) and all patients were of White race. The mean duration of epilepsy at the time of implantation was 70 ± 32 months (range, 9-166 months). Based on the current clinical guidelines in Russia, all patients were evaluated for epilepsy surgery prior to consideration of VNS Therapy.
All registry patients had received VNS Therapy for at least 2 years or more per the entry criteria for the registry. The analysis included follow-up data ranging from 24 to 101 months after implantation.
As listed in Table 1, the majority of registry patients were diagnosed with focal epilepsy (n = 29; 50%). Other types of epilepsy were combined focal and generalized epilepsy (n = 12; 20%) and generalized epilepsy (n = 8; 14%). The epilepsy type was unknown in nine patients (16%).
Assigning causation in a heterogeneous disorder such as epilepsy can be complicated, and more than one epilepsy etiology was reported for some of the patients. Overall, the epilepsy etiologies reported by the patients included genetic disorders (n = 15; 26%), structural disorders (n = 14; 24%), and infectious causes (n = 2; 3%), and epileptic seizures in 27 patients (47%) were of unknown etiology.
A total of 49 patients experienced focal seizures (85%) and nine patients experienced generalized seizures (15%). The various types of seizures are listed in Table 2.
Most of the patients (98%) were receiving one to four ASMs during the time of the VNS implantation. Six of the 58 patients had also received nonpharmacological treatments prior to implantation including three patients with resective surgery, two patients had received ketogenic diet, and one patient had palliative cranial surgery (callosotomy).

Changes in frequency of the most disabling seizure types
The changes in the frequency and severity of the "most disabling" type of epileptic seizures (defined as seizures that were accompanied with falls, physical trauma, and/or incontinence in the absence of preventative measures) were assessed for the baseline levels (at implantation) compared with the available data from 24 months and any timepoint later than 24 months postimplantation.
Of the 58 pediatric patients, 48 had a history of experiencing the most disabling seizure type and an additional patient who did not experience disabling seizures at baseline reported such seizures 2 years after starting VNS Therapy. Overall, these patients experienced a significant decrease in the frequency of the most disabling seizure type following VNS Therapy from a mean of 29.8 seizures per month at implantation to a mean of 7.2 seizures per month after 24 months of treatment (p < .001) (Figure 1). Notably at 24 months after implantation, 16 of the 49 patients (33%) reported that they TA B L E 1 Baseline demographics, disease characteristics, and ASM use in the registry patients who received VNS Therapy for at least 2 years.  (Figure 2a).
Five patients (10%) reported an increase in the frequency of such seizures (including the patient mentioned above who did not have disabling seizures before implantation but reported them 2 years after implantation).
TA B L E 2 Seizure classification in the pediatric patients in the registry.

Changes in frequency and severity of all seizure types based on McHugh Outcome scale
Based on the McHugh Outcome scale, patient outcomes were classified into five levels of seizure reduction (Classes I-V). Classes I, II, and III were further subdivided into A or B according to the effects in relation to ictal and/or postictal severity.
As presented in Figure 2b, the frequency dynamics of all types of epileptic seizures with 24 months of VNS Therapy resulted in 18 patients (31%) with a Class I outcome (80%-100% reduction in seizure frequency), 19 patients (33%) with a Class II outcome (50%-79% reduction), and 10 patients (17%) with a Class III outcome (<50% reduction), and two patients (3%) had a Class IV outcome (benefited only from the external magnet). The remaining nine patients (16%) had a Class V outcome (i.e., did not experience any improvements in seizure frequency and no magnet effects) and of these, four patients (7%) reported worsening symptoms.
Importantly, 11 of the patients with a Class I outcome experienced a complete remission (i.e., were free of all types of seizures). Therefore, 19% of the analysis population achieved remission.

Seizure severity
The questionnaire to assess seizure severity was completed by 43 patients who participated in an additional study visit after 24 months following implantation. A final score was generated based on the data on seizure-related factors, and the results demonstrated that the severity of epileptic seizures and postseizure states in pediatric patients with adjunctive VNS Therapy decreased by 44%.

Status epilepticus
Status epilepticus is a common neurological emergency in pediatric patients. Therefore, we assessed the presence of status epilepticus defined as a seizure with 5 min or more of continuous clinical and/or electrographic seizure activity or recurrent seizure activity without recovery between seizures (Wylie et al., 2022). Status epilepticus was noted in 13 patients (22%) prior to implantation with a mean number of 9.4 ± 17.7 (range, 0-52) status epilepticus per year in these patients.
After implantation, status epilepticus was reported in only one patient with an average of about 1 event per 4-6 months.

VNS parameters
We studied the effect of the stimulation parameters on the effectiveness of VNS Therapy, including output current (mA), signal frequency The mean, the standard deviation of the mean, and the minimummaximum range of various parameters are presented in Table 4.
ANOVA was utilized to evaluate VNS Therapy parameters (output current, ON time, OFF time, duty cycle, pulse frequency, and pulse width). The analysis showed that there were significant differences only for ON time (p = .004) among groups with different seizure fre-quency outcomes compared with the control visit (i.e., a visit that took place any time following 24 months after implantation) (Table 5).
Tukey's HSD test indicated significant difference in mean among the seizure freedom group and the group with 50%−99% decrease in seizure frequency (p = .005).

Safety findings
Device-or treatment-related adverse events included the following events in one patient (1.7%) each: hoarseness at 3 months; hoarseness (n = 1) and paresthesia and cough (n = 1) at 6 months; cough at 12 months; hoarseness and cough at 24 months; and cough was reported at a control visit more than 24 months after implantation.
There were no serious adverse events or adverse reactions that led to discontinuation of VNS Therapy.

DISCUSSION
Here, we have reported the efficacy and safety results from a multicenter retrospective observational registry involving pediatric patients with DRE who were treated with VNS Therapy for at least 2 years.
Several articles have been published presenting use of VNS Therapy in Russian patients with DRE based on either limited patient samples or case studies (Aivazyan et al., 2016;Areshkina et al., 2019;Katyshev et al., 2020;Pylaeva et al., 2019). To our knowledge, our study represents the first multicenter study to evaluate the safety and efficacy of treatment with adjunctive VNS Therapy in pediatric patients with DRE across the Russian Federation.
To ensure a robust evaluation process, changes in seizure frequency following implantation were evaluated for the "most disabling" epileptic seizures (defined as seizures accompanied by falls, physical trauma, and/or incontinence in the absence of preventative measures) and changes in seizure frequency and severity were evaluated for all epileptic seizure types.
About 26% of our study patients were diagnosed with genetic epilepsies. This rate appears to be higher than that in studies of VNS in a DRE population that included children and adults (Elliott et al., 2011), but it is aligned with the findings from Symonds and colleagues who identified genetic etiologies in 31% of pediatric patients under 3 years of age (Symonds et al., 2021). VNS Therapy is programed by clinicians using an external programing device to select the right combination of adjustable parameters (output current, signal frequency, pulse width, and signal ON and OFF times) to reach the best antiseizure effects while minimizing stimulation-induced side effects. Therefore, one of the aims of our study was to evaluate the response rates in the Russian Federation compared with studies conducted in other countries. The estimated rate of 19% of patients who experienced seizure freedom in our study is higher than the pooled prevalence estimates of 12% identified in a meta-analysis of 101 clinical studies evaluating VNS Therapy in pediatric patients (Jain & Arya, 2021). But the proportion of patients with ≥50% decrease in frequency of all seizure types in our study was 64%, which is higher than the pooled prevalence estimates of 56% based on other studies of pediatric patients (Jain & Arya, 2021) but in alignment with other published studies (Elliott et al., 2011;Rossignol et al., 2009). The higher positive outcome in our study may have arisen from only including patients who had achieved ≥1.5 mA output current within 9 months of implantation, and the results require further investigation.

CONCLUSION
VNS Therapy is an effective and safe adjunct treatment for DRE in male and female pediatric patients.