Characterization of white matter over 1–2 years in small vessel disease using MR-based quantitative susceptibility mapping and free-water mapping

Yawen Sun, Ying Hu, Yage Qiu, Yuyao Zhang, Changhao Jiang, Peiwen Lu, Qun Xu, Yuting Shi, Hongjiang Wei, Yan ZhouFrontiers in Aging Neuroscience 2022 · DOI

TL;DR

This longitudinal study reveals that in Cerebral Small Vessel Disease (SVD), the "normal-appearing" white matter (NAWM) is a site of active, progressive injury even when visible lesions remain stable. By using advanced MRI techniques like Quantitative Susceptibility Mapping (QSM) and Free-Water (FW) mapping, researchers identified that axonal loss in the prefrontal lobe serves as a critical early biomarker for cognitive decline.

What Problem Does This Paper Address?

Cerebral Small Vessel Disease (SVD) is a major contributor to stroke and the leading cause of vascular dementia. Clinically, SVD is often tracked via White Matter Lesions (WMLs) visible on standard MRI. However, these lesions represent late-stage damage—the "tip of the iceberg."

The real challenge lies in the "Normal-Appearing White Matter" (NAWM). While it looks healthy on standard scans, microstructural degradation is often already underway. Standard Diffusion Tensor Imaging (DTI) struggles to distinguish between different types of tissue damage (like demyelination vs. inflammation) because of "free-water" contamination from cerebrospinal fluid or edema. This paper aims to use multi-parametric MRI to characterize how NAWM changes over 1–2 years and how those changes correlate with cognitive performance.

Methods

The researchers followed 51 SVD patients over a 1–2 year period, performing neuropsychological testing and advanced 3.0T MRI at both baseline and follow-up. They employed two primary advanced techniques:

  • Quantitative Susceptibility Mapping (QSM): This measures the magnetic susceptibility of tissue. In white matter, this is highly sensitive to myelin content; an increase in susceptibility (moving toward more positive values) serves as an index for demyelination.
  • Free-Water (FW) Mapping: This uses a bi-tensor model to separate the diffusion properties of water molecules within tissue from those in the extracellular space. This allows for:
  • FW fraction: An index of neuroinflammation or edema.
  • FW-corrected DTI ($FA_T$ and $MD_T$): More accurate measures of actual tissue integrity (e.g., axonal loss) stripped of the "noise" caused by extracellular fluid.

The team used the JHU-ICBM-DTI atlas to segment specific white matter tracts for a granular, ROI-based analysis.

Datasets

The study utilized a primary longitudinal dataset of 51 SVD patients recruited from the Neurology Department of Ren Ji Hospital, Shanghai.

  • Inclusion criteria: Age 50–85, presence of subcortical lacunar infarcts/WMLs, and a modified Rankin score ≤ 3.
  • Anatomic Guide: The researchers used the standard JHU-ICBM-DTI label atlas to ensure consistent anatomical segmentation across the cohort.

Key Insights and Ideas

  • NAWM is the Active Battleground: While visible WMLs remained relatively stable in volume and composition over the study period, the NAWM showed significant microstructural changes. This suggests that SVD is a chronic, progressive "whole-brain" disease, not just a collection of isolated spots.
  • The Prefrontal Connection: The most significant changes occurred in the prefrontal lobe (specifically the superior and inferior frontal blades). Because the prefrontal lobe is central to executive function and attention, damage here is a "smoking gun" for early cognitive impairment.
  • Demyelination Precedes Axonal Loss: Increased QSM values (demyelination) were widespread across the NAWM at follow-up, whereas significant decreases in $FA_T$ (axonal loss) were more localized. This suggests a sequence of degradation that advanced imaging can track.
  • Inflammation Matters: FW analysis revealed increased extracellular water in the right corticospinal tract and inferior frontal blade, indicating that neuroinflammation or vasogenic edema is a dynamic component of SVD progression.

Key Results

  • Cognitive Stability: Interestingly, overall cognitive scores (MMSE, MoCA) did not show a significant decline over the 1–2 year period, likely due to active risk factor management (hypertension/diabetes control).
  • Susceptibility Shift: QS values in the NAWM increased significantly ($p = 0.029$), specifically in the left superior frontal blade, left occipital blade, and right uncinate fasciculus.
  • Axonal Loss vs. Cognition: A decrease in $FA_T$ (axonal loss) in the right inferior frontal blade was positively correlated with changes in MoCA ($r = 0.365, p = 0.012$) and MMSE ($r = 0.309, p = 0.034$) scores. This identifies axonal loss in this specific region as a potential biomarker for cognitive evolution.
  • WML Stability: No significant changes were found in any MRI indices within the existing white matter lesions, reinforcing that NAWM is the more sensitive area for longitudinal monitoring.

Limitations and Future Work

  • Sample Size and Control Group: The study was limited to 51 patients and lacked a healthy age-matched control group, making it difficult to separate SVD progression from normal aging.
  • Short Duration: The 1–2 year window may be too short to capture gross cognitive decline or major lesion expansion.
  • Statistical Correction: The authors noted that multiple comparison corrections were not applied, as this was an exploratory study.
  • Future Directions: The authors suggest larger, longer-term multicenter studies and exploring how these white matter changes relate to gray matter atrophy to provide a more complete picture of SVD-related dementia.

Summary updated

Abstract

The NAWM was still in the progressive injury process over time, while WMLs remained relatively stable, which supports the notion that SVD is a chronic progressive disease. The process of axonal loss in the NAWM part of the prefrontal lobe might be a biomarker of cognitive changes in the evolution of SVD.
BibTeX
@inproceedings{sun2022characterization,
  author = {Yawen Sun and Ying Hu and Yage Qiu and Yuyao Zhang and Changhao Jiang and Peiwen Lu and Qun Xu and Yuting Shi and Hongjiang Wei and Yan Zhou},
  title = {{Characterization of white matter over 1–2 years in small vessel disease using MR-based quantitative susceptibility mapping and free-water mapping}},
  booktitle = {Frontiers in Aging Neuroscience},
  year = {2022},
  doi = {10.3389/fnagi.2022.998051},
}
Figure