Circuit Decoding Across Scales: DBS and EEG Integration in Parkinson's Disease for Motor and Non-Motor Symptoms
DOI: Pending DOI
Synopsis
Non-motor symptoms, including mood disturbances and pathological impulsivity, significantly impact quality of life in patients with Parkinson's disease (PD), yet neural signatures linking DBS target selection to those aspects are still undetermined. We recorded high-density EEG during rest and movement control in PD patients with STN or GPi DBS (ON/OFF states), extracting beta-related features including coherence and phase-amplitude coupling. The Beck Depression Inventory and the Barratt Impulsiveness Scale assessed non-motor symptoms. Results on the first cohort reveal target-specific cortical signatures: STN-DBS modulates broader prefrontal networks implicated in impulse control, while GPi effects are sensorimotor-restricted. Neural measures correlated with impulsivity scores, supporting the development of EEG biomarkers for personalized neuromodulation of mood and cognitive symptoms.
Background
Non-motor symptoms—including depression, anxiety, and impulse control disorders—affect up to 60% of Parkinson's disease (PD) patients and often impact quality of life even more than motor disturbances (1,2). Deep brain stimulation (DBS) target selection may differentially influence these outcomes: STN-DBS has been associated with changes in impulsivity and mood fluctuations, whereas GPi-DBS may carry lower cognitive and neuropsychiatric risks (3,4). However, neural biomarkers predicting individual cognitive and psychiatric responses to each target remain lacking.
STN and GPi engage distinct basal ganglia-cortical circuits. STN-DBS modulates the hyperdirect pathway with strong prefrontal projections implicated in impulse regulation, while GPi-DBS influences direct/indirect pathways with more restricted sensorimotor effects (5,6). Scalp EEG can characterize these differential cortical signatures through beta oscillations and phase-amplitude coupling (PAC), features implicated in both motor and cognitive-affective control.
We aimed to identify EEG biomarkers linking DBS target to non-motor symptom profiles by: (1) comparing cortical effects of STN versus GPi DBS; (2) examining relationships between neural signatures and depression (Beck Depression Inventory) and impulsivity (Barratt Impulsiveness Scale). This approach addresses the clinical need for biomarkers guiding target selection based on individual psychiatric profiles.
Methods
Participants: PD patients with implanted DBS targeting STN (n=4) or GPi (n=3) were recruited, with ongoing enrollment toward n=15 per group plus 15 matched healthy controls.
Clinical Assessments: Beck Depression Inventory-II (BDI-II) assessed depressive symptoms across cognitive-affective and somatic domains. Barratt Impulsiveness Scale-11 (BIS-11) assessed trait impulsivity across attentional, motor, and nonplanning subscales—dimensions relevant to impulse control disorders reported with STN-DBS.
Protocol: 64-channel scalp EEG was recorded during DBS-ON and DBS-OFF conditions (randomized, 30-minute washout). Sessions included: (1) 5-minute eyes-open resting state; (2) continuous movement task (6 blocks, 108 trials) engaging frontal-basal ganglia circuits relevant to impulse regulation.
EEG Analysis: Data were bandpass filtered (0.5-200 Hz), notch filtered at stimulation frequency, and cleaned via ICA. Features extracted: (1) beta burst characteristics (13-30 Hz); (2) beta-gamma phase-amplitude coupling; (3) frontal-sensorimotor beta coherence; (4) waveform shape. Correlations between EEG features and BDI/BIS scores were examined.
Results
Impulsivity (BIS-11):
Total impulsivity scores showed a trend toward higher values in STN versus GPi patients (STN mean = 68.0, range 59-74; GPi mean = 62.3, range 49-70), consistent with literature associating STN-DBS with impulse control changes. Subscale analysis revealed that nonplanning impulsiveness—reflecting difficulty with future orientation and self-control—showed the greatest target differentiation (STN: 26.3; GPi: 24.0). Motor impulsiveness was similar across groups (STN: 24.7; GPi: 23.0). Notably, one STN patient (score=74) exceeded the clinical threshold for high impulsivity (>72), warranting monitoring for impulse control disorders.
Depression (BDI-II):
Both groups showed minimal total depressive symptoms (STN mean = 7.0; GPi mean = 7.0). However, individual item analysis revealed clinically meaningful variability in somatic-performance symptoms: fatigue (Q17: 67% endorsed), sleep disturbance (Q16: 67%), and work difficulty (Q15: 83%). These subsyndromal symptoms, common in PD, may represent targets for neuromodulation even when total depression scores remain low.
Target-Specific Cortical Signatures:
EEG revealed distinct spatial patterns of DBS effects aligned with psychiatric-relevant circuits. During DBS-ON versus OFF:
STN group: Beta power reduction extended beyond the sensorimotor cortex to the prefrontal regions, including sites overlying the dorsolateral prefrontal cortex implicated in impulse control. Frontal-sensorimotor beta coherence was substantially reduced. Phase-amplitude coupling (PAC) modulation was similarly extended to prefrontal electrodes.
GPi group: Beta modulation was more spatially restricted to the sensorimotor cortex, with minimal prefrontal effects. Long-range coherence changes were less pronounced than STN.
These patterns suggest STN-DBS engages broader frontal networks governing cognitive-affective functions, while GPi-DBS effects remain sensorimotor-focused—potentially explaining differential psychiatric outcomes between targets.
Brain-Behavior Correlations (n=7):
Frontal beta coherence correlated with BIS-11 motor impulsiveness subscale: higher frontal-sensorimotor coherence was associated with lower impulsivity scores, suggesting this connectivity metric may index prefrontal control over impulsive responding. BDI somatic items (fatigue, work difficulty) correlated with resting sensorimotor beta power, indicating even subsyndromal mood symptoms have neural correlates detectable via scalp EEG.
Discussion
Our findings reveal target-specific cortical signatures with direct relevance to psychiatric and cognitive symptom profiles in PD. STN-DBS modulates prefrontal networks implicated in impulse control, consistent with higher impulsivity scores observed in STN patients and clinical reports of impulse control disorders following STN stimulation. GPi-DBS effects appear sensorimotor-restricted, potentially explaining its less complicated cognitive and mood-related impact.
The correlation between frontal beta coherence and impulsivity scores supports this metric as a candidate biomarker for risk stratification of mood and impulsivity disorders. Patients showing greater prefrontal modulation with DBS may warrant closer monitoring for impulse control changes, while those with minimal prefrontal effects may tolerate higher stimulation intensities without psychiatric complications.
Even with minimal depression scores, somatic symptoms (fatigue, sleep, work difficulty) correlated with sensorimotor beta activity, suggesting EEG biomarkers may track clinically relevant non-motor symptoms below diagnostic thresholds. This has implications for optimizing DBS to address quality-of-life symptoms beyond motor function.
Our next endeavor will integrate Percept™ subcortical recordings to correlate STN/GPi local field potentials with these cortical signatures, enabling closed-loop approaches targeting both psychiatric and motor outcomes.
Disclosures
This work was supported by NIH R01 NS134772-01.
Nothing to disclose by any author(s).
References
1) Aarsland D, Marsh L, Schrag A. Neuropsychiatric symptoms in Parkinson's disease. Mov Disord. 2009;24(15):2175-2186.
2) Schrag A, Jahanshahi M, Quinn N. What contributes to quality of life in patients with Parkinson's disease? J Neurol Neurosurg Psychiatry. 2000;69(3):308-312.
3) Antonini A, Siri C, Santangelo G, et al. Impulsivity and compulsivity in drug-naïve patients with Parkinson's disease. Mov Disord. 2011;26(3):464-468.
4) Follett KA, Weaver FM, Stern M, et al. Pallidal versus subthalamic deep-brain stimulation for Parkinson's disease. N Engl J Med. 2010;362(22):2077-2091.
5) Li Q, Ke Y, Chan DC, et al. Therapeutic deep brain stimulation in parkinsonian rats directly influences motor cortex. Neuron. 2012;76(5):1030-1041.
6) Gunalan K, Howell B, McIntyre CC. Quantifying axonal responses in patient-specific models of subthalamic deep brain stimulation. Neuroimage. 2017;172:263-277.
7) Swann NC, de Hemptinne C, Aron AR, et al. Elevated synchrony in Parkinson disease detected with electroencephalography. Ann Neurol. 2015;78(5):742-750.
8) Parmigiani S, Mikulan E, Russo S, et al. Simultaneous stereo-EEG and high-density scalp EEG recordings to study the effects of intracerebral stimulation parameters. Brain Stimul. 2022;15(2):372-375.