Basal ganglia theta power indexes trait anxiety in people with Parkinson's disease

Abstract

Neuropsychiatric symptoms are common and disabling in Parkinson's disease, with troublesome anxiety occurring in one-third of patients. Management of anxiety in Parkinson's disease is challenging, hampered by insufficient insight into underlying mechanisms, lack of objective anxiety measurements and largely ineffective treatments. In this study, we assessed the intracranial neurophysiological correlates of anxiety in patients with Parkinson's disease treated with deep brain stimulation (DBS) in the laboratory and at home. We hypothesized that low-frequency (theta-alpha) activity would be associated with anxiety. We recorded local field potentials from subthalamic nucleus or globus pallidus pars interna DBS implants in three Parkinson's disease cohorts: (i) patients with recordings (subthalamic nucleus) performed in hospital at rest via perioperatively externalized leads, without active stimulation, both ON and OFF dopaminergic medication; (ii) patients with recordings (subthalamic nucleus or globus pallidus pars interna) performed at home while resting, via a chronically implanted commercially available sensing-enabled neurostimulator (Medtronic Percept™ device), ON dopaminergic medication, with stimulation both on and off; and (iii) patients with recordings performed at home while engaging in a behavioural task via subthalamic nucleus and globus pallidus pars interna leads and electrocorticography paddles over the premotor cortex connected to an investigational sensing-enabled neurostimulator, ON dopaminergic medication, with stimulation both on and off. Trait anxiety was measured with validated clinical scales in all participants, and state anxiety was measured with momentary assessment scales at multiple time points in the two at-home cohorts. Power in theta (4-8 Hz) and alpha (8-12 Hz) ranges was extracted from the local field potential recordings, and its relationship with anxiety ratings was assessed using linear mixed-effects models. In total, 33 patients with Parkinson's disease (59 hemispheres) were included. Across three independent cohorts, with stimulation off, basal ganglia theta power was positively related to trait anxiety (all P < 0.05). Also in a naturalistic setting, with individuals at home, at rest, with stimulation and medication ON, basal ganglia theta power was positively related to trait anxiety (P < 0.05). This relationship held regardless of the hemisphere and DBS target. There was no correlation between trait anxiety and premotor cortical theta-alpha power. There was no within-patient association between basal ganglia theta-alpha power and state anxiety. We showed that basal ganglia theta activity indexes trait anxiety in Parkinson's disease. Our data suggest that theta could be a possible physiomarker of neuropsychiatric symptoms and specifically of anxiety in Parkinson's disease, potentially suitable for guiding advanced DBS treatment tailored to the needs of the individual patient, including non-motor symptoms.

Keywords: Parkinson's disease; anxiety; deep brain stimulation; local field potential; non-motor symptoms.

Figure 1

Three independent investigational cohorts for neurophysiological analysis. This study included three cohorts of Parkinson’s disease patients treated with deep brain stimulation. In the ‘in-hospital externalized at-rest’ cohort, basal ganglia LFP recordings were obtained from the STN a few days after surgery via externalized electrodes with the patient at rest. Patients were either OFF or ON medication, but stimulation was not yet enabled. Anxiety was rated with the Hamilton Anxiety Rating Scale (HARS). In the ‘at-home chronic at-rest’ cohort, basal ganglia (STN or GPi) LFPs were obtained at home in an unsupervised but semi-structured setting using the Percept PC's BrainSense technology (BrainSense ‘Event’ triggered by the patient) with the patient at rest. Stimulation was either off or on, but patients were always in a medication ON state. Anxiety was rated with the Beck Anxiety Inventory (BAI). Raw LFPs were not available in the ‘at-home chronic at-rest’ cohort, whereas in the other two cohorts offline LFP processing was required to obtain frequency-domain data. In the ‘at-home chronic task’ cohort, basal ganglia (STN or GPi) and cortical (premotor cortex) LFPs were obtained via remotely supervised streaming via the Summit RC+S system with the patient at rest between trials of a behavioural task. Experimental conditions involved medication ON–stimulation on and medication ON–stimulation off. Anxiety was rated with the BAI. In all three cohorts, analyses were focused on the relationship between anxiety measures and power in the theta and alpha frequency bands. Infographics in the first row have been generated with the help of the generative artificial intelligence platform Stable Diffusion Online (https://stablediffusionweb.com/). LFP = local field potential; GPi = globus pallidus pars interna; PSD = power spectral density; STN = subthalamic nucleus.

Figure 2

Anxiety is related to basal ganglia theta in in-hospital externalized recordings. This figure displays the relationship between anxiety measured by the HARS and basal ganglia (STN) theta and alpha band power in the ‘in-hospital externalized at-rest’ cohort. Each colour represents a patient (n = 12 patients), and empty/filled markers indicate right/left hemisphere recordings (n = 24 hemispheres). (A) With medication OFF–stimulation off, a positive relationship is present between anxiety and basal ganglia theta. (B–D) No relationship between anxiety and theta is present with medication ON-stimulation off; no relationship between anxiety and alpha in any condition. (E) Coefficients of anxiety in LME models. *P < 0.05. CI = confidence interval; HARS = Hamilton anxiety rating scale; LME = linear mixed-effects; STN = subthalamic nucleus.

Figure 3

Anxiety is related to basal ganglia theta and alpha in at-home at-rest recordings. Relationship between anxiety and subcortical LFP in the ‘at-home chronic at-rest’ cohort. Scatter plot colours indicate the participants (n = 13 patients), with circles indicating STN implants and squares GPi implants. Empty/filled markers indicate right/left hemisphere recordings (n = 19 hemispheres). Theta and alpha band power were extracted from at-home at-rest BrainSense Event LFP recordings, and correlation was assessed with baseline anxiety measured with the BAI. With medication ON–stimulation off (A and C) and medication ON–stimulation on (B and D), anxiety is positively related to subcortical theta (A and B) and alpha (C and D) band power. (E) Coefficients of BAI in LME models. *P < 0.05. BAI = Beck Anxiety Inventory; CI = confidence interval; GPi = globus pallidus pars interna; LFP = local field potential; LME = linear mixed-effects model; STN = subthalamic nucleus.

Figure 4

Anxiety is related to basal ganglia theta in at-home recordings during a cognitive task. Relationship between anxiety and subcortical LFP during a reward task in the ‘at-home chronic task’ cohort. Theta and alpha power were extracted from the intertrial interval in a reward task, and correlation was assessed with baseline anxiety (BAI). Scatter plot colours indicate participants (n = 8 patients), with circles indicating STN implants and squares GPi implants. Empty/filled markers indicate right/left hemisphere recordings (n = 15 hemispheres). With medication ON–stimulation off (A), there is a positive relationship between anxiety and subcortical theta band power. (B–D) No relationship between anxiety and theta is present with medication ON-stimulation on; no relationship between anxiety and alpha in any condition. (E) Coefficients of BAI in LME models. *P < 0.05. BAI = Beck Anxiety Inventory; CI = confidence interval; GPi = globus pallidus pars interna; LFP = local field potential; LME = linear mixed-effects model; STN = subthalamic nucleus.