Evidence of White Matter Neuroinflammation in Myalgic Encephalomyelitis/Chronic Fatigue Syndrome: A Diffusion‐Based Neuroinflammation Imaging Study

Qiang Yu, Kiana Kothe, Richard A. Kwiatek, Peter Del Fante, Anya Bonner, Vince D. Calhoun, Zack Y. ShanHuman 2026 · DOI

TL;DR

This study uses a novel diffusion MRI technique called Neuroinflammation Imaging (NII) to provide the first in vivo evidence of white matter neuroinflammation in Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/CFS). The findings suggest that NII metrics, unlike traditional Diffusion Tensor Imaging (DTI), can detect subtle inflammatory changes and correlate with clinical symptoms, potentially serving as valuable biomarkers for ME/CFS.

What Problem Does This Paper Address?

Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/CFS) is a debilitating illness with suspected neuroinflammatory underpinnings, but its biological mechanisms remain poorly understood. Previous neuroimaging studies using standard Diffusion Tensor Imaging (DTI) have yielded inconsistent results regarding white matter abnormalities in ME/CFS patients. This inconsistency might stem from DTI's limitations in specifically identifying inflammatory processes, as it can be confounded by edema or cellular infiltration. The authors aim to address this by employing an advanced diffusion-based model, Neuroinflammation Imaging (NII), which is designed to disentangle these processes and provide more specific markers of neuroinflammation in the white matter of ME/CFS patients.

Methods

The core of this study is the application of the Neuroinflammation Imaging (NII) model. Unlike conventional DTI, which models water diffusion using a single tensor, NII models the diffusion MRI signal as a combination of anisotropic diffusion tensors and an isotropic diffusion spectrum. This allows for the estimation of several biologically informed indices:

  • NII-HR (Hindered Water Ratio): Reflects extracellular tissue edema.
  • NII-RF (Restricted Fraction): Indicates inflammation-related cellularity.
  • NII-FF (Fibre Fraction): Represents apparent axonal density.
  • NII-AD, NII-RD, NII-MD, NII-FA: Fibre-compartment axial, radial, mean diffusivity, and fractional anisotropy, respectively.

These NII metrics were computed from diffusion MRI data. Tract-Based Spatial Statistics (TBSS) was used for voxel-wise group comparisons between ME/CFS patients and healthy controls (HCs). Additionally, multiple regression analyses were performed to investigate the relationships between NII/DTI metrics and various clinical measures (mental health, physical health, sleep quality, disability, disease severity, and duration).

Datasets

The study utilized diffusion MRI data from a cohort of 67 ME/CFS patients and 67 rigorously matched healthy controls (HCs). These participants were drawn from a previously described cohort (Yu et al., 2025). The demographic and clinical data for these participants are presented in Table 1. No new datasets were introduced; the study leverages existing data with a novel analytical approach.

Key Insights and Ideas

  • NII is more sensitive than DTI for ME/CFS: The study highlights that conventional DTI metrics showed minimal group differences and no significant clinical associations, while the NII model revealed widespread abnormalities and significant correlations with clinical symptoms. This underscores the added value of NII in detecting subtle neuroinflammatory changes.
  • ME/CFS exhibits a distinct neuroinflammatory signature: The findings suggest that ME/CFS is characterized by cerebral edema (reduced NII-HR), cellular infiltration (reduced NII-RF), and axonal reorganisation (increased NII-FF) in white matter. This provides a more specific picture of the underlying pathology than previous DTI studies.
  • Neuroinflammation is linked to clinical symptoms: Crucially, the study found that reduced NII-RF (indicating less cellular infiltration) was associated with worse mental health and higher disability levels in ME/CFS patients. This directly links the observed neuroinflammatory markers to the lived experience of the illness.
  • Heterogeneity of ME/CFS might be reflected in diffusion patterns: While the study didn't stratify by onset type for the main findings, the detection of both increased and decreased NII-AD and NII-MD suggests a complex and potentially heterogeneous pattern of white matter changes.

Key Results

  • Group Differences: ME/CFS patients showed significantly lower NII-HR (edema) and NII-RF (cellular infiltration) and significantly higher NII-FF (axonal density) across various white matter tracts compared to HCs. Some regions also showed decreased NII-AD and NII-MD.
  • Clinical Associations:
    • Lower NII-RF was associated with worse mental health (lower MCS) and increased disability (higher BDS) across all participants.
  • Lower NII-AD and NII-MD were associated with worse mental health (lower MCS) in ME/CFS patients.
  • Higher NII-FF was associated with lower disease severity in ME/CFS patients.
  • DTI vs. NII: Conventional DTI metrics showed minimal group differences and no significant clinical associations, reinforcing the superiority of NII for this investigation.

Limitations and Future Work

The authors acknowledge several limitations:

  • Indirect Measure: NII is an indirect measure of neuroinflammation and cannot differentiate between specific inflammatory cell types or processes.
  • Lack of Biological Samples: The study did not collect biological samples, preventing the investigation of peripheral biomarkers that could be associated with the observed brain changes.
  • Future Directions: The authors suggest future research should incorporate validation with other neuroinflammation-specific techniques and biological sampling to improve clinical translation and facilitate subtype-specific approaches. They also highlight the potential of NII-derived metrics as biomarkers for ME/CFS.

Summary updated

Abstract

Myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) is a debilitating disorder with suspected neuroinflammatory pathophysiology. However, previous diffusion tensor imaging (DTI) studies have reported inconsistent white matter abnormalities in ME/CFS, and specific white matter inflammatory changes remain poorly characterised. This study employed an advanced diffusion-based neuroinflammation imaging (NII) model to investigate white matter neuroinflammation in ME/CFS. Diffusion MRI data from 67 ME/CFS patients (median age, 38; and 54 women) and 67 rigorously matched healthy controls (HCs) (median age 38; and 52 women) were analysed. Seven NII-derived metrics were computed: hindered water ratio (NII-HR), restricted fraction (NII-RF), fibre fraction (NII-FF), axial diffusivity (NII-AD), radial diffusivity (NII-RD), mean diffusivity (NII-MD) and fractional anisotropy (NII-FA). Conventional DTI metrics were also calculated. Tract-based spatial statistics were used to perform voxel-wise group comparisons, and multiple regression analysis was conducted to examine the relationship between NII/DTI metrics and clinical measures of mental health, physical health, sleep quality, disability, disease severity and disease duration. Compared to HCs, ME/CFS patients exhibited widespread white matter abnormalities, including significantly lower NII-HR and NII-RF, and significantly higher NII-FF, NII-AD, NII-MD and NII-FA across association, commissural and projection fibres. Additionally, some regions showed decreased NII-AD and NII-MD in ME/CFS. Lower NII-RF, NII-AD and NII-MD in ME/CFS were significantly associated with worse mental health, while lower NII-RF was also associated with a higher level of disability. Among ME/CFS patients, higher NII-FF was associated with lower disease severity. Conventional DTI showed minimal group differences and no significant clinical associations. This study provides in vivo evidence of white matter neuroinflammation in ME/CFS, characterised by cerebral edema (reduced NII-HR), cellular infiltration (reduced NII-RF) and axonal reorganisation (increased NII-FF). This suggests NII-derived indices may serve as sensitive biomarkers for neuroinflammation in ME/CFS.
BibTeX
@inproceedings{yu2026evidence,
  author = {Qiang Yu and Kiana Kothe and Richard A. Kwiatek and Peter Del Fante and Anya Bonner and Vince D. Calhoun and Zack Y. Shan},
  title = {{Evidence of White Matter Neuroinflammation in Myalgic Encephalomyelitis/Chronic Fatigue Syndrome: A Diffusion‐Based Neuroinflammation Imaging Study}},
  booktitle = {Human},
  year = {2026},
  doi = {10.1002/hbm.70505},
}
Figure