Neurite orientation and dispersion density imaging (NODDI) detects cortical and corticospinal tract degeneration in ALS

Rebecca Broad, Matt C. Gabel, Nicholas G. Dowell, David J. Schwartzman, Anil K. Seth, Hui Zhang, Daniel C. Alexander, Mara Cercignani, P. Nigel LeighJournal of Neurology Neurosurgery & Psychiatry 2018 · DOI

Abstract

Neurite orientation dispersion and density imaging (NODDI) demonstrates that axonal loss within the corticospinal tract (CST) is a core feature of degeneration in Amyotrophic Lateral Sclerosis (ALS). This axonal loss is the primary factor contributing to the altered diffusivity profile typically detected using standard Diffusion Tensor Imaging (DTI). Furthermore, NODDI identified dendritic alterations within the precentral gyrus (PCG), suggesting that microstructural cortical dendritic changes occur concurrently with CST axonal damage.

Introduction

Amyotrophic Lateral Sclerosis (ALS) is a devastating neurodegenerative disease characterized by the progressive loss of motor neurons in the brain and spinal cord. Clinical assessment often relies on MRI to track these changes; however, standard techniques like Diffusion Tensor Imaging (DTI) possess significant limitations. DTI measures the movement of water molecules to infer tissue microstructure, but its primary metric, Fractional Anisotropy (FA), lacks specificity. A drop in FA could indicate either a loss of axonal density or a change in fiber organization (e.g., fanning or crossing). This study addresses the need for "in vivo" histology to distinguish between these processes and determine whether ALS pathology is primarily a "dying back" (axonal) or "dying forward" (cortical) phenomenon.

Methods

The study compared 23 patients with sporadic ALS against 23 age-matched healthy controls. Imaging was performed on a 1.5T Siemens AVANTO scanner. The researchers utilized NODDI, a multi-compartment diffusion model that partitions the MRI signal into three environments: intra-neurite water (axons/dendrites), extra-neurite water, and cerebrospinal fluid (CSF). From these, three indices were derived:

  1. Neurite Density Index (NDI): Quantifies the volume fraction of axons and dendrites.
  2. Orientation Dispersion Index (ODI): Measures the complexity or "fanning" of neurite architecture.
  3. Isotropic Fraction (ISO): Measures free water, often reflecting tissue atrophy.

These metrics were compared against standard DTI parameters (FA and Mean Diffusivity) using whole-brain voxel-wise analysis.

Results

The primary finding was that the reduction in FA observed in ALS is driven predominantly by a decrease in NDI. This confirms that actual axonal loss, rather than simple disorganization, is the central feature of white matter degeneration in the CST.

Figure 3: NODDI maps showing significant reductions in NDI (A), ODI (B), and increases in ISO (C) in ALS patients compared to controls.

NODDI demonstrated significantly higher sensitivity than DTI. The cluster size for reduced NDI was approximately three times larger than that for reduced FA (29,349 vs. 9,283 voxels), capturing extensive transcallosal involvement that DTI missed.

Figure 4: Standard DTI maps showing reduced FA (A) and MD (B) in the CST and pons, illustrating the more limited detection compared to NODDI.

Furthermore, the study identified reduced ODI in the right precentral gyrus (motor cortex). This suggests that cortical atrophy in ALS involves a specific loss of dendritic arborization. Clinical correlations revealed that both ODI and ISO correlated with disease duration, indicating that the motor cortex becomes less complex and more fluid-filled as the disease progresses. Subgroup analysis also showed that patients with bulbar involvement had significantly lower NDI in the corona radiata compared to those with limb-only symptoms.

Discussion and Limitations

While NODDI provides a more granular view of ALS pathology, the study has limitations. The use of a 1.5T scanner, while compensated for by larger voxel sizes, lacks the spatial resolution of 3T or 7T systems. The sample size of 23 patients is relatively small given the heterogeneity of ALS phenotypes. Additionally, the cross-sectional design prevents definitive conclusions about the predictive value of NODDI for disease progression. Future longitudinal studies and the use of more advanced models like Bingham-NODDI could further refine our understanding of directional fiber fanning.

Conclusion

NODDI represents a significant advancement over standard DTI for mapping ALS pathology. By decoupling neurite density from orientation, it reveals that axonal loss is the primary driver of white matter changes and identifies concurrent cortical dendritic degeneration, providing a more comprehensive "in vivo" histological profile of the disease.

Summary updated

Abstract

NODDI demonstrates that axonal loss within the CST is a core feature of degeneration in ALS. This is the main factor contributing to the altered diffusivity profile detected using DTI. NODDI also identified dendritic alterations within the PCG, suggesting microstructural cortical dendritic changes occur together with CST axonal damage.
BibTeX
@article{broad2018neurite,
  author = {Rebecca Broad and Matt C. Gabel and Nicholas G. Dowell and David J. Schwartzman and Anil K. Seth and Hui Zhang and Daniel C. Alexander and Mara Cercignani and P. Nigel Leigh},
  title = {{Neurite orientation and dispersion density imaging (NODDI) detects cortical and corticospinal tract degeneration in ALS}},
  journal = {Journal of Neurology Neurosurgery \& Psychiatry},
  year = {2018},
  doi = {10.1136/jnnp-2018-318830},
}
Figure