Intracranial beta activity is a biomarker of circadian and stimulation-induced arousal in obsessive compulsive disorder

Abstract

Deep brain stimulation (DBS) has been investigated as therapy for treatment-resistant neuropsychiatric disorders, such as obsessive-compulsive disorder (OCD) [[1]Denys D. Graat I. Mocking R. et al.Efficacy of deep brain stimulation of the ventral anterior limb of the internal capsule for refractory obsessive-compulsive disorder: a clinical cohort of 70 patients.Am J Psychiatr. 2020; 177: 265-271https://doi.org/10.1176/appi.ajp.2019.19060656Crossref PubMed Scopus (91) Google Scholar] and major depressive disorder (MDD) [[2]Dougherty D.D. Rezai A.R. Carpenter L.L. et al.A randomized sham-controlled trial of deep brain stimulation of the ventral capsule/ventral striatum for chronic treatment-resistant depression.Biol Psychiatr. 2015; 78: 240-248https://doi.org/10.1016/J.BIOPSYCH.2014.11.023Abstract Full Text Full Text PDF PubMed Scopus (0) Google Scholar]. However, side effects associated with increased arousal, such as insomnia and manic symptoms [[3]Widge A.S. Licon E. Zorowitz S. et al.Predictors of hypomania during ventral capsule/ventral striatum deep brain stimulation.J Neuropsychiatry Clin Neurosci. 2016; 28: 38-44https://doi.org/10.1176/appi.neuropsych.15040089Crossref Scopus (33) Google Scholar,[4]Alonso P. Cuadras D. Gabriëls L. et al.Deep brain stimulation for obsessive-compulsive disorder: a meta-analysis of treatment outcome and predictors of response.PLoS One. 2015; 10e0133591https://doi.org/10.1371/journal.pone.0133591Crossref Scopus (221) Google Scholar], can limit the efficacy of therapy. Manic symptoms, which include a decreased need for sleep, have been reported in up to 30–45 % of individuals receiving anterior limb of internal capsule (ALIC) stimulation [[3]Widge A.S. Licon E. Zorowitz S. et al.Predictors of hypomania during ventral capsule/ventral striatum deep brain stimulation.J Neuropsychiatry Clin Neurosci. 2016; 28: 38-44https://doi.org/10.1176/appi.neuropsych.15040089Crossref Scopus (33) Google Scholar] and require urgent care. Insomnia is also a frequently reported adverse side effect [[4]Alonso P. Cuadras D. Gabriëls L. et al.Deep brain stimulation for obsessive-compulsive disorder: a meta-analysis of treatment outcome and predictors of response.PLoS One. 2015; 10e0133591https://doi.org/10.1371/journal.pone.0133591Crossref Scopus (221) Google Scholar], characterized by an inability to transition from wakefulness to sleep or frequent sleep disruptions. Furthermore, disruptions in sleep-wake function increase the risk of developing mania. While the underlying mechanisms of these side effects remain unclear, many of the activating effects of ALIC DBS can be reversed by decreasing the amplitude of stimulation [[3]Widge A.S. Licon E. Zorowitz S. et al.Predictors of hypomania during ventral capsule/ventral striatum deep brain stimulation.J Neuropsychiatry Clin Neurosci. 2016; 28: 38-44https://doi.org/10.1176/appi.neuropsych.15040089Crossref Scopus (33) Google Scholar]. These clinical observations indicate that ALIC DBS may recruit striato-limbic regions that modulate the baseline tone of arousal. Emerging literature suggests that incorporating the arousal state [[5]Scangos K.W. Makhoul G.S. Sugrue L.P. Chang E.F. Krystal A.D. State-dependent responses to intracranial brain stimulation in a patient with depression.Nat Med. 2021; 27: 229-231https://doi.org/10.1038/s41591-020-01175-8Crossref PubMed Scopus (96) Google Scholar] and/or circadian rhythms into the stimulation approach may improve treatment and minimize related side effects [[6]Gilron R. Little S. Wilt R. Perrone R. Anso J. Starr P.A. Sleep-aware adaptive deep brain stimulation control: chronic use at home with dual independent linear discriminate detectors.Front Neurosci. 2021; 15732499https://doi.org/10.3389/fnins.2021.732499Crossref Scopus (14) Google Scholar]. A better understanding of the electrophysiological biomarkers of circadian and stimulation-related arousal could enable the development of adaptive ALIC DBS systems. Cortical arousal manifests as shifts in activity towards higher frequencies. Increased beta activity measured from scalp EEG during wake and sleep has been associated with insomnia [[7]Zhao W. Van Someren E.J.W. Li C. et al.EEG spectral analysis in insomnia disorder: a systematic review and meta-analysis.Sleep Med Rev. 2021; 59101457https://doi.org/10.1016/j.smrv.2021.101457Crossref Scopus (68) Google Scholar]. Furthermore, beta activity within intracranial EEG electrodes has been demonstrated to capture sleep-wake states across cortical sites [[8]Fan J.M. Lee A.M. Sellers K.K. et al.Intracranial electrical stimulation of corticolimbic sites modulates arousal in humans.Brain Stimul. 2023; 16: 1072-1082https://doi.org/10.1016/j.brs.2023.06.017Abstract Full Text Full Text PDF Scopus (0) Google Scholar]. In applications for Parkinson's disease, intracranial beta activity within subcortical regions shows circadian variability [[6]Gilron R. Little S. Wilt R. Perrone R. Anso J. Starr P.A. Sleep-aware adaptive deep brain stimulation control: chronic use at home with dual independent linear discriminate detectors.Front Neurosci. 2021; 15732499https://doi.org/10.3389/fnins.2021.732499Crossref Scopus (14) Google Scholar,[9]Yin Z. Ma R. An Q. et al.Pathological pallidal beta activity in Parkinson's disease is sustained during sleep and associated with sleep disturbance.Nat Commun. 2023; 14: 5434https://doi.org/10.1038/s41467-023-41128-6Crossref Scopus (0) Google Scholar]. We therefore hypothesized that beta activity within striato-limbic regions may be a circadian biomarker of arousal. In a cohort of four treatment resistant OCD patients receiving DBS therapy, we aimed to better understand the mechanism of ALIC DBS-induced arousal by performing intracranial recordings from the neighboring striato-limbic region (Fig. 1A). We aimed to determine: 1) if striato-limbic beta activity is an electrophysiological biomarker of circadian variation of arousal and 2) whether DBS alters local striato-limbic beta in a manner that tracks with stimulation-induced arousal. We recorded continuous band-limited spectral power within the beta frequency range in an ambulatory setting for 6–84 days (Table S1; Supplementary methods). These chronic recordings showed circadian fluctuations over long time scales, as in example patient OCD 1 (Fig. 1B and C). The mean periodogram confirmed a circadian peak at 24 hrs (Fig. 1D). Across all subjects and recording sites, the detrended striato-limbic beta power depicted a significant change across a 24-hour clock (Fig. 1E and F; Tables S2 and S3 for individual statistical testing). Across leads, the sinusoidal fit was statistically significant (Table S2; p < 0.0001 in all leads) and mean percent of variation explained by time of day was 11.1 % [SD 3.73 %] (Fig. 1G, Table S2). The detrended data was also converted to polar coordinates to quantify the non-uniform distribution of beta power (Fig. 1F, Table S3). The magnitude of resultant vectors was significant (Table S3; p < 0.05) compared to shuffled data in all subjects except the left lead from OCD 3. We next assessed the impact of changes in stimulation amplitude on arousal level in two of the subjects (OCD 1 and 4). Arousal levels were measured by the SSS and VAS-energy scales as stimulation was cycled from low to high amplitudes (Supplementary methods). Spectrograms were computed from LFP recordings from both the stimulation lead (example patient, left lead, Fig. 1H) and contralateral sensing lead. Across both subjects, an increase in stimulation amplitude was associated with an increase in subjective energy levels (two-sample Wilcoxon rank sum test, p = 0.006; Fig. 1I), but not sleepiness levels (Fig. S1A, p = 0.348). The linear combination of the two behavioral responses represented by the dominant principal component remained statistically significant with respect to stimulation amplitude (p = 0.043; Fig. S2A). Increased stimulation amplitude corresponded to an acute increase in beta spectral power in the stimulation lead (Fig. S1B, left panel, p < 0.001), but not the sensing lead (Fig. S1B, right panel). Furthermore, DBS-induced changes in arousal measured with the VAS-energy showed significant positive correlation with striato-limbic beta power in both the stimulation lead (Fig. 1J, Pearson's correlation, r = 0.522, p = 0.005) and sensing lead (Fig. 1K, r = 0.407, p = 0.035). Incorporating SSS levels in tandem with the VAS-energy as a combined measure of arousal using the dominant principal component improved the correlation with beta power in the sensing lead (Fig. S2C, r = 0.461, p = 0.016), suggesting that both measures may capture relevant behavioral measures that reflect DBS-induced changes in spectral power. To our knowledge, this is the first study characterizing the circadian variation of beta activity within the striato-limbic region. It is likely that other frequency bands may also reflect a circadian pattern [[8]Fan J.M. Lee A.M. Sellers K.K. et al.Intracranial electrical stimulation of corticolimbic sites modulates arousal in humans.Brain Stimul. 2023; 16: 1072-1082https://doi.org/10.1016/j.brs.2023.06.017Abstract Full Text Full Text PDF Scopus (0) Google Scholar]. Prior work has demonstrated that regions neighboring the ALIC, such as the BNST and striatum, have key roles in circadian regulation [[10]Giardino W.J. Eban-Rothschild A. Christoffel D.J. Li S Bin Malenka R.C. de Lecea L. Parallel circuits from the bed nuclei of stria terminalis to the lateral hypothalamus drive opposing emotional states.Nat Neurosci. 2018; 21 (2018 218): 1084-1095https://doi.org/10.1038/s41593-018-0198-xCrossref PubMed Scopus (140) Google Scholar], which may explain the strong circadian variation. Arousal effects stemming from ALIC stimulation may be related to local effects on striato-limbic regions. Prior studies have suggested that ALIC DBS can release dopamine [[11]Figee M. De Koning P. Klaassen S. et al.Deep brain stimulation induces striatal dopamine release in obsessive-compulsive disorder.Biol Psychiatr. 2014; 75: 647-652https://doi.org/10.1016/j.biopsych.2013.06.021Abstract Full Text Full Text PDF PubMed Scopus (85) Google Scholar], which could help sustain states of wakefulness. Alternatively, stimulation of white matter fibers may indirectly activate connected regions. For example, the anterior thalamic radiations connect the prefrontal cortex to the thalamus [[12]Li N. Baldermann J.C. Kibleur A. et al.A unified connectomic target for deep brain stimulation in obsessive-compulsive disorder.Nat Commun. 2020; 11https://doi.org/10.1038/S41467-020-16734-3Crossref Google Scholar]. Our study demonstrates two key features of DBS targeting the ALIC region that may enable an adaptive, closed-loop paradigm to account for arousal effects: a biomarker and a tunable parameter of stimulation that modulates arousal. Incorporation of arousal effects into a closed-loop DBS system may mitigate stimulation-induced hyperarousal side effects, including sleep disruption and manic symptoms. In this framework, the striato-limbic beta biomarker may be used as a feedback signal to tailor stimulation amplitude to natural circadian variation of arousal. For instance, decreasing stimulation when excessive beta activity is detected during nighttime hours may prevent nighttime hyperarousal, which contributes to insomnia and mania. Primary sleep-wake disorders, such as excessive daytime sleepiness, could also be treated with closed-loop approaches that promote natural circadian variation of arousal. This work is supported by the National Institute of Mental Health (Grant No. R21MH130914, K23MH125018), National Institute of Neurological Disorders and Stroke (Grant No. K23NS125123), P&S Fund from the Brain and Behavior Foundation (Grant No. A136828), Ray and Dagmar Dolby Family Fund, and Foundation for OCD Research (Grant No. P0548058).

Topics & Concepts