Authors :
Presenting Author: John Ingham, BSc, MB BS, MSc – Newcastle University
Frances Hutchings, PhD – Newcastle University; Sadegh Soudjani, PhD – Newcastle University; Yujiang Wang, PhD – Newcastle University; Paolo Zuliani, PhD – Newcastle University; Peter Taylor, PhD – Newcastle University
Rationale:
Vagal nerve stimulation (VNS) is used for medication-resistant epilepsy, and can reduce seizure frequency and duration in some patients[1]. The VNS stimulation protocol is open-loop and adjusted for each patient over successive clinic visits, based on previous clinical experience and by trial and error. The precise mechanisms of VNS are incompletely understood. Computational models may aid understanding and can be used to make patient-specific predictions. Here we propose such a model, which, to our knowledge, has not previously been achieved.
Methods:
We take as our starting point an established, four-population neural mass model that can reproduce seizure dynamics in a thalamocortical circuit [2]. Noise added to the system produces transitions between seizing and normal states. We then extend the model to include 18 further neuronal populations representing 11 brain regions relevant to VNS. VNS stimulation is simulated by altering the background input to the excitatory and inhibitory populations of the nucleus tractus solitarius (NTS), which receives inputs from the vagus nerve in vivo. Bifurcation analysis was performed on the model to determine the effect of varying inputs to the NTS. We further compared model simulations with and without stimulation under identical noise conditions to investigate the effect of VNS.
Results:
Bifurcation analyses showed that VNS tended to move the model from a bistable system, with a fixed point (representing normal activity) and limit cycle (representing a seizure state), to a monostable system with a fixed point only. Simulated VNS caused some seizures to terminate early, and for other simulations seizures did not develop at all, thus replicating the effect of VNS in silico.
Conclusions:
We successfully produced a biologically plausible in silico model of VNS in epilepsy, capturing behavior seen in vivo. This may aid in our understanding of the therapeutic mechanisms of VNS in epilepsy and provide a starting point to (i) determine which patients are likely to respond best to VNS, and (ii) optimize individualized patient treatments, ultimately with the possible addition of closed-loop control.
References
[1] Morris GL 3rd, Mueller WM. Long-term treatment with vagus nerve stimulation in patients with refractory epilepsy. The Vagus Nerve Stimulation Study Group E01-E05. Neurology. 1999 Nov 10;53(8):1731-5. doi: 10.1212/wnl.53.8.1731. Erratum in: Neurology 2000 Apr 25;54(8):1712. PMID: 10563620
[2] Taylor, P. N., Wang, Y., Goodfellow, M., Dauwels, J., Moeller, F., Stephani, U., Baier, G. A computational study of stimulus driven epileptic seizure abatement. PLOS one, 9(12), e114316. 2014
Funding:
This work was supported by the Engineering and Physical Sciences Research Council [Grant Number 2595464].
P.N.T. and Y.W. are both supported by UKRI Future Leaders Fellowships (MR/T04294X/1, MR/V026569/1).