Acute Brain Activation Patterns of High- Versus Low-Frequency Stimulation of the Anterior Nucleus of the Thalamus During Deep Brain Stimulation for Epilepsy

Abstract

BACKGROUND Deep brain stimulation (DBS) of the anterior nucleus of the thalamus (ANT) is an increasingly utilized treatment of drug-resistant epilepsy. To date, the effect of high-frequency stimulation (HFS) vs low-frequency stimulation (LFS) in ANT DBS is poorly understood. OBJECTIVE To assess differences in the acute effect of LFS vs HFS in ANT DBS utilizing blood-oxygen-level-dependent (BOLD) functional magnetic resonance imaging (fMRI). METHODS In this prospective study of 5 patients with ANT DBS for epilepsy, BOLD activation and deactivation were modeled for 145-Hz and 30-Hz ANT stimulation using an fMRI block design. Data were analyzed with a general linear model and combined via 2-stage mixed-effects analysis. Z-score difference maps were nonparametrically thresholded using cluster threshold of z > 3.1 and a (corrected) cluster significance threshold of P = .05. RESULTS HFS produced significantly greater activation within multiple regions, in particular the limbic and default mode network (DMN). LFS produced minimal activation and failed to produce significant activation within these same networks. HFS produced widespread cortical and subcortical deactivation sparing most of the limbic and DMN regions. Meanwhile, LFS produced deactivation in most DMN and limbic structures. CONCLUSION Our results show that HFS and LFS produce substantial variability in both local and downstream network effects. In particular, largely opposing effects were identified within the limbic network and DMN. These findings may serve as a mechanistic basis for understanding the potential of HFS vs LFS in various epilepsy syndromes.

Deep brain stimulation targets in epilepsy: Systematic review and meta‐analysis of anterior and centromedian thalamic nuclei and hippocampus

“…34,52 A recent study with low-and high-frequency stimulation during resting state fMRI showed that HFS produces activation of DMN and limbic networks of the brain in contrast to LFS. 53 Herrman et al 13 concluded that low response to ANT DBS in their study could have been associated with using a fixed set of electrodes and parameters in all patients without individual adjustment. Open-source research software (Lead-DBS) and commercial software Scientific STIMVIEW) are available that can estimate the field of stimulation, which may improve adjustment of stimulation parameters by targeting brain regions associated with successful treatment (ie, proximity to MTT).…”

Probing responses to deep brain stimulation with functional magnetic resonance imaging

“…ANT stimulation produced activation within several areas of the brain corresponding to the default mode and limbic networks, including the thalamus, bilateral anterior cingulate and posterior cingulate cortex, precuneus, medial prefrontal cortex, amygdala, ventral tegmental area, hippocampus, striatum, and right angular gyrus. Furthermore, Sarica et al and Middlebrooks et al showed that these areas were not activated when the locus of stimulation was moved away from the ANT and when low-frequency stimulation (30 Hz) was used, further highlighting the potential for fMRI in selecting optimal stimulation parameters [58,59].…”

Identifying the neural network for neuromodulation in epilepsy through connectomics and graphs

“… 23 A recent study with low- and high-frequency stimulation during resting state fMRI showed that high frequency DBS produced activation of DMN and limbic networks of the brain, in contrast to low-frequency stimulation. 22 Studies of functional connectivity associated with CMT electrode VTAs produced less functional connectivity in regions of DMN, but higher connectivity with sensorimotor, premotor, ACC, ascending reticular activating system (ARAS) and other brain regions ( Table 2 ). 18 , 19 …”

“…The common functional connectivity network shared by clinically used DBS targets overlapped regions of the epileptic networks identified by previous studies of regions activated during DBS. 14 , 15 , 22 , 23 This suggests that DBS targets and epilepsy share a common cortico-subcortical network that might be responsible for the antiseizure action of DBS. 21 , 46 , 47 …”