Abstracts

Rationale: Vagus nerve stimulation (VNS) has been a FDA approved therapy for epilepsy since 1997 for patients 12 and older with treatment resistant partial onset seizures. Efficacy and safety in younger pediatric patients have been seen with use of VNS therapy as well. Adult studies have demonstrated reduction in emergency department (ED) visits and hospitalizations which correlated with estimated and also actual cost savings following VNS implantation. However, a newer model of VNS was released in June 2016. The new device has a tachycardia detection system to active the device with an increase heart rate as programed by the medical provider team. No studies have been performed determining the effect of the new device on healthcare utilization. Therefore, we performed an analysis utilizing the electronic health record (EHR) to measure the intervention of VNS therapy in regards to its effect on healthcare utilization. Methods: Institutional Review Board (IRB) approval was obtained for this study. The study design was a retrospective data analysis of the electronic health record (EHR) data from Nationwide Children's Hospital for all patients who had received the AspireSR 106 VNS device. Patients with a current procedural terminology (CPT) code (64568) for initial VNS placement was searched in the claims database from June 2016 to June 2017. A counts analysis was performed and compared the number of visits of ED, unplanned inpatient admissions, and neurology encounters within 9-12 months before and after an initial VNS placement. Unplanned inpatient admissions were defined as inpatient visits resulting from the ED. Nine months was selected give additional patients to query. Descriptive analyses were performed to characterize patients’ age and gender. Wilcoxon signed rank test was conducted to evaluate the changes in the average number of unplanned inpatient, ED and neurology visits within 9-12 months before and after the patient had a VNS implantation. All analyses were conducted using SAS 9.4 (SAS Institute, Cary, NC). Two-sided P values < .05 were considered statistically significant. Results: The data search identified 34 unique patients with VNS placement during June 2016 and June 2017 with at least nine months of post implantation data. The range of ages for patients was 4.6 to 34.4 years of age. The average age at time of implantation was 15.4 years of age. About 55.9% of the study population were male (N=19) and 44.1% were female (N=15). Unplanned hospitalizations dropped by 34.3% by then end of post-VNS data collection. ED visits dropped 33.9% by the end of post-VNS data collection. Anti-seizure medications prescribed dropped by 7.5% by the end of post-VNS data collection. Calls to nursing triage dropped by 14.0% by the end of post-VNS data collection. Conclusions: Utilizing electronic health record data (EHR), VNS implantation with the AspireSR model demonstrates a significant reduction in unplanned hospitalizations, ED visits, and calls to the neurology triage nurses. The reduction noted also will have a decreased financial impact as a result of decreased unplanned hospitalizations and ED visits. Our case series is novel in that we evaluated data on healthcare utilization for children and adults getting the AspireSR 106 VNS model which has the heart rate detection. Limitations of our study are implicit in using a small data set from one institution for analysis. In addition, we were unable to calculate the costs associated with VNS implantation as that data was not available in the EHR. Our analysis does not account for possible surgical complications that may have occurred resulting in higher healthcare utilization immediately following VNS implantation, although our numbers did not reflect this tendency. In addition, we only evaluated one center and due to the recent release of this device only 9 months of data can be evaluated at this time. Further collection of the data at 1 and 2 years to substantiate these findings would be helpful. In addition, similar analysis in a larger cohort in many geographic locations would add to the possibility if similar results were noted. In summary, our study demonstrates a decrease in healthcare utilization following VNS implantation with the newest VNS device, AspireSR, in a free standing children’s hospital in Columbus, Ohio. Given the burden of inpatient hospitalizations for treatment resistant children with epilepsy, VNS implantation may be an effective method of treatment to decrease this burden and costs associated with the hospitalizations. Funding: This study received an investigator initiated grant from LivaNova.