Magnetoencephalography detects phase-amplitude coupling in Parkinson’s disease

Masataka Tanaka, Takufumi Yanagisawa, Ryohei Fukuma, Naoki Tani, Satoru Oshino, Masahito Mihara, Noriaki Hattori, Yuta Kajiyama, Ryota Hashimoto, Manabu Ikeda, Hideki Mochizuki, Haruhiko KishimaScientific 2022 · DOI

Abstract: Abnormal phase-amplitude coupling (PAC) in the cortico-basal ganglia circuit is a known hallmark of Parkinson's disease (PD), but its whole-brain distribution has remained largely unexplored due to the invasive nature of traditional recording methods. This study utilizes non-invasive magnetoencephalography (MEG) to characterize PAC across the entire cortex in patients with PD compared to healthy controls. Results reveal that PD patients exhibit significant beta-gamma PAC widely distributed across sensorimotor, occipital, and temporal regions. Critically, this coupling in the sensorimotor cortex correlates with the severity of motor symptoms (akinesia), whereas traditional measures like spectral power do not. These findings establish beta-gamma PAC as a robust, non-invasive biomarker for Parkinsonian pathology and symptom severity.

Introduction

Parkinson’s disease (PD) is fundamentally a disorder of brain rhythm, characterized by pathological neural oscillations within the circuits connecting the cortex and the basal ganglia. A prominent feature of this dysfunction is Phase-Amplitude Coupling (PAC)—a phenomenon where the phase of a low-frequency oscillation (typically in the beta band, 13–25 Hz) modulates the power of a high-frequency oscillation (gamma band, 50–100 Hz).

Until now, most PAC research in PD required invasive surgery to place electrodes deep in the brain (DBS) or directly on the cortical surface (ECoG). While effective, these methods are limited to surgical candidates and provide a narrow view of brain activity. This study addresses the need for a high-resolution, whole-brain, non-invasive method to track PD pathology using Magnetoencephalography (MEG).

Methods

The study recorded resting-state MEG signals from 23 patients with PD and 23 age- and sex-matched healthy study participants (HSPs). To achieve high spatial resolution, magnetic signals were projected onto individual 3D cortical models using "minimum-norm estimation" and the FreeSurfer atlas. The cortex was further divided into 360 distinct areas using the Human Connectome Project (HCP) parcellation for standardized analysis.

The primary metric used was the Synchronization Index (SI), which quantifies how consistently the timing of the beta-band phase aligns with the peaks of the gamma-band amplitude. These neural metrics were then correlated against the MDS-UPDRS-III akinesia scores, a standard clinical measure of motor impairment.

Results and Spatial Distribution

The whole-brain MEG approach revealed that abnormal beta–gamma PAC in PD is distributed far more widely than previously thought. While previous invasive studies focused on the motor cortex, significant coupling was found across the sensorimotor, occipital, and temporal cortices in PD patients.

Comparison of Beta-gamma PAC (SI values) between PD patients (A) and healthy participants (B), showing the widespread distribution of coupling in the patient group.

In contrast, healthy participants showed only minimal coupling in small parts of the somatosensory and temporal regions. Statistical analysis confirmed that beta–gamma PAC in the sensorimotor cortex significantly correlated with akinesia scores ($r = 0.45, P = 0.02$). Notably, traditional measures like beta-band power ($r = 0.04$) and gamma-band power ($r = -0.13$) showed no such correlation, highlighting PAC as a superior biomarker for motor symptoms.

Key Insights

  • Widespread Pathology: This is the first report of significant beta–gamma PAC in the visual (occipital) and temporal regions of PD patients. The researchers suggest this widespread synchronization might eventually explain non-motor symptoms like impaired visual perception or hallucinations.
  • PAC vs. Power: The study found that while the "loudness" (power) of specific brain waves might differ between groups, it is the coupling between frequencies that serves as a reliable indicator of clinical impairment.
Spatial distribution of significant Beta-gamma PAC t-values, illustrating the areas of highest statistical significance in the brain.

Limitations and Future Directions

Most participants were recorded in their "ON" medication state. Since dopamine replacement therapy is known to suppress PAC, the study likely underestimated the true strength of the abnormal coupling. Future research in the "OFF" state could reveal even stronger biomarkers. Additionally, while PAC correlates with symptoms, it remains unclear if it is a cause of the symptoms or a byproduct of the underlying pathology. Currently, this MEG protocol serves as a powerful tool for monitoring symptom severity and treatment efficacy rather than a standalone diagnostic test.

Conclusion

By demonstrating that non-invasive MEG can detect and map abnormal beta-gamma PAC, this research provides a new window into the whole-brain dynamics of Parkinson’s disease. The correlation between sensorimotor PAC and akinesia establishes a clear link between neural synchronization and clinical symptoms, offering a promising path for non-invasive disease monitoring.

Summary updated

Abstract

To characterize Parkinson's disease, abnormal phase-amplitude coupling is assessed in the cortico-basal circuit using invasive recordings. It is unknown whether the same phenomenon might be found in regions other than the cortico-basal ganglia circuit. We hypothesized that using magnetoencephalography to assess phase-amplitude coupling in the whole brain can characterize Parkinson's disease. We recorded resting-state magnetoencephalographic signals in patients with Parkinson's disease and in healthy age- and sex-matched participants. We compared whole-brain signals from the two groups, evaluating the power spectra of 3 frequency bands (alpha, 8-12 Hz; beta, 13-25 Hz; gamma, 50-100 Hz) and the coupling between gamma amplitude and alpha or beta phases. Patients with Parkinson's disease showed significant beta-gamma phase-amplitude coupling that was widely distributed in the sensorimotor, occipital, and temporal cortices; healthy participants showed such coupling only in parts of the somatosensory and temporal cortices. Moreover, beta- and gamma-band power differed significantly between participants in the two groups (P < 0.05). Finally, beta-gamma phase-amplitude coupling in the sensorimotor cortices correlated significantly with motor symptoms of Parkinson's disease (P < 0.05); beta- and gamma-band power did not. We thus demonstrated that beta-gamma phase-amplitude coupling in the resting state characterizes Parkinson's disease.
BibTeX
@inproceedings{tanaka2022magnetoencephalography,
  author = {Masataka Tanaka and Takufumi Yanagisawa and Ryohei Fukuma and Naoki Tani and Satoru Oshino and Masahito Mihara and Noriaki Hattori and Yuta Kajiyama and Ryota Hashimoto and Manabu Ikeda and Hideki Mochizuki and Haruhiko Kishima},
  title = {{Magnetoencephalography detects phase-amplitude coupling in Parkinson’s disease}},
  booktitle = {Scientific},
  year = {2022},
  doi = {10.1038/s41598-022-05901-9},
}
Figure