Serial deep gray nuclear DTI changes in Parkinson’s disease over twelve years

Yao‐Chia Shih, Leon Qi Rong Ooi, Huihua Li, John Carson Allen, Septian Hartono, Thomas Welton, Eng‐King Tan, Ling Ling ChanFrontiers in Aging Neuroscience 2023 · DOI

TL;DR

This study presents a rare 12-year longitudinal look at how Parkinson’s Disease (PD) reshapes the deep gray matter of the brain. By tracking patients over three timepoints, researchers found that Mean Diffusivity (MD) in the caudate nucleus progressively increases and correlates strongly with clinical motor decline, suggesting it could serve as a valuable objective biomarker for tracking late-stage disease progression.

What Problem Does This Paper Address?

While we know Parkinson’s Disease involves the progressive loss of dopaminergic neurons, mapping the exact "in vivo" trajectory of neurodegeneration in the basal ganglia over a decade is incredibly difficult. Most existing studies using Diffusion Tensor Imaging (DTI)—a technique that measures the movement of water molecules to infer tissue microstructure—are either cross-sectional (a single snapshot) or short-term (1–2 years).

Because PD is a chronic, decades-long condition, short-term studies often produce inconsistent results. There is a critical need for long-term data to understand how different brain regions—like the caudate, putamen, and thalamus—degenerate at different rates and how these changes relate to worsening motor symptoms.

Methods

The researchers conducted a case-control study following 149 subjects (72 PD patients and 77 healthy controls) over 12 years, with MRI scans and clinical assessments every 6 years.

  • Imaging: All subjects were scanned on the same 1.5T MRI scanner using a standardized DTI protocol to ensure consistency.
  • Metrics: They focused on two primary DTI indices: Fractional Anisotropy (FA), which measures the directionality of water diffusion (often linked to structural integrity), and Mean Diffusivity (MD), which measures the overall magnitude of diffusion (linked to tissue density and neurodegeneration).
  • Analysis: They used a semi-automated pipeline to segment four key regions: the caudate, putamen, globus pallidus, and thalamus. Statistical analysis employed multivariate linear mixed-effects regression models to account for aging, gender, and the inevitable "dropout" of participants over a 12-year period.

Datasets

The study utilized a longitudinal cohort from the Singapore General Hospital.

  • Baseline: 149 subjects (72 PD / 77 HC).
  • Year 6: 97 subjects returned.
  • Year 12: 49 subjects remained. The researchers also analyzed a "subset cohort" of the 44 individuals who completed all three timepoints to ensure that the results weren't skewed by the participants who dropped out.

Key Insights and Ideas

  • MD is a clearer signal than FA: While FA is often the "go-to" metric in DTI, this study found FA changes in the basal ganglia to be "complex" and sometimes non-linear. MD, however, showed a steady, progressive increase in PD patients, making it a more reliable indicator of cumulative neurodegeneration.
  • The Caudate as a Late-Stage Tracker: Interestingly, while the putamen and globus pallidus showed early changes, Caudate MD was the metric that most consistently correlated with clinical severity (UPDRS-III and Hoehn and Yahr scores). This suggests the caudate might be the "canary in the coal mine" for tracking the transition into advanced PD.
  • The Iron Confound: The authors highlight a fascinating biological "noise" factor: iron accumulation. In healthy aging, iron builds up in the putamen, which can artifactually increase FA. In PD, the loss of tissue integrity (which lowers FA) competes with this iron effect, explaining why FA results in previous studies have been so inconsistent.

Key Results

  • Putaminal and Pallidal MD: Significantly higher in PD patients compared to controls by Year 12 ($p < 0.001$ and $p = 0.002$, respectively).
  • Caudate MD Correlation: MD in the caudate showed a strong positive correlation with disease duration ($p < 0.0001$) and clinical motor scores ($p < 0.05$).
  • FA Trajectory: Thalamic FA actually increased at Year 6 in PD patients (possibly a compensatory response) before dropping or stabilizing, while putaminal FA significantly decreased by Year 12 ($p = 0.0189$).

Limitations and Future Work

  • Field Strength: The study used 1.5T MRI. While this provided longitudinal consistency, modern 3T or 7T scanners offer higher resolution and better signal-to-noise ratios.
  • The "Free Water" Problem: Standard DTI assumes a simple model of water diffusion. The authors suggest that future work should use more advanced models (like Diffusion Kurtosis Imaging or bi-tensor models) to better separate the effects of neurodegeneration from "free water" (fluid) increases.
  • Iron Quantification: Because iron deposition significantly affects DTI metrics in the basal ganglia, the authors recommend that future longitudinal studies include specific iron-sensitive sequences (like QSM) to "de-noise" the DTI data.

Summary updated

Abstract

Pallido-putaminal MD showed differential neurodegeneration in PD over 12 years on longitudinal DTI; putaminal and thalamic FA changes were complex. Caudate MD could serve as a surrogate marker to track late PD progression.
BibTeX
@inproceedings{shih2023serial,
  author = {Yao‐Chia Shih and Leon Qi Rong Ooi and Huihua Li and John Carson Allen and Septian Hartono and Thomas Welton and Eng‐King Tan and Ling Ling Chan},
  title = {{Serial deep gray nuclear DTI changes in Parkinson’s disease over twelve years}},
  booktitle = {Frontiers in Aging Neuroscience},
  year = {2023},
  doi = {10.3389/fnagi.2023.1169254},
}
Figure