CoChat
Sign Up
Cited by 291Advanced Neuroimaging Techniques and ApplicationsMultiple Sclerosis Research StudiesAdvanced MRI Techniques and Applications

Neurite dispersion: a new marker of multiple sclerosis spinal cord pathology?

Open access
Grussu, Francesco, Schneider, Torben, Tur, Carmen, Yates, Richard L., Tachrount, Mohamed, Ianuş, Andrada, Yiannakas, Marios C., Newcombe, Jia +4 more·2017·Annals of Clinical and Translational Neurology
DOI 10.1002/acn3.445PMC PMC5590517
Open this paper in CoChat — free accountOpen PDF

Abstract

Objective: Conventional magnetic resonance imaging ( MRI ) of the multiple sclerosis spinal cord is limited by low specificity regarding the underlying pathological processes, and new MRI metrics assessing microscopic damage are required. We aim to show for the first time that neurite orientation dispersion (i.e., variability in axon/dendrite orientations) is a new biomarker that uncovers previously undetected layers of complexity of multiple sclerosis spinal cord pathology. Also, we validate against histology a clinically viable MRI technique for dispersion measurement ( neurite orientation dispersion and density imaging, NODDI ), to demonstrate the strong potential of the new marker. Methods: We related quantitative metrics from histology and MRI in four post mortem spinal cord specimens (two controls; two progressive multiple sclerosis cases). The samples were scanned at high field, obtaining maps of neurite density and orientation dispersion from NODDI and routine diffusion tensor imaging ( DTI ) indices. Histological procedures provided markers of astrocyte, microglia, myelin and neurofilament density, as well as neurite dispersion. Results: We report from both NODDI and histology a trend toward lower neurite dispersion in demyelinated lesions, indicative of reduced neurite architecture complexity. Also, we provide unequivocal evidence that NODDI ‐derived dispersion matches its histological counterpart ( P < 0.001), while DTI metrics are less specific and influenced by several biophysical substrates. Interpretation: Neurite orientation dispersion detects a previously undescribed and potentially relevant layer of microstructural complexity of multiple sclerosis spinal cord pathology. Clinically feasible techniques such as NODDI may play a key role in clinical trial and practice settings, as they provide histologically meaningful dispersion indices.

AI snapshot

Written from the paper’s full text.

TL;DR

In four post mortem spinal cord specimens (two controls, two progressive multiple sclerosis cases), neurite orientation dispersion measured by NODDI and histology showed a trend toward lower dispersion in demyelinated lesions. NODDI-derived dispersion matched its histological counterpart, supporting neurite dispersion as a potentially specific marker of multiple sclerosis spinal cord microstructural pathology.

Key findings

  • Both NODDI orientation dispersion index (ODI) and histological circular variance (CV) showed a trend toward lower neurite orientation dispersion in demyelinated lesions, indicating reduced neurite architecture complexity.
  • NODDI ODI was strongly associated with histological CV in controls (r = 0.84; P < 0.001) and multiple sclerosis cases (r = 0.60; P = 0.001); linear regression also showed ODI–CV association in controls (β = 0.860, P < 10−5) and multiple sclerosis (β = 0.711, P < 10−4).
  • DTI metrics were less specific: they were influenced by demyelination, neuroaxonal loss, and glial changes, and no DTI metric showed association with histology-derived dispersion (CV), unlike NODDI ODI.
  • NODDI neurite density index (NDI) was sensitive to neuronal density but also strongly influenced by myelin, limiting interpretability without myelin mapping.

Methods

  • Ex vivo study relating quantitative MRI and histology in four formalin-fixed post mortem spinal cord specimens: two nonneurological controls and two progressive multiple sclerosis cases.
  • Specimens were scanned at 9.4 T with diffusion MRI; NODDI maps (IVF, NDI, ODI) and DTI indices (FA, AD, RD, MD) were derived. Histology provided CV, myelin staining fraction (MSF), neurofilament staining fraction (NSF), astrocyte staining fraction (ASF), and microglia staining fraction (μGSF).
  • MRI and histology were co-registered; ROI-wise medians were extracted for N = 48 ROIs (28 multiple sclerosis, 20 controls). Pearson correlations and univariate multivariable linear regression models were used, with MRI metrics as dependent variables and histological indices as predictors.

Limitations

  • Paper-stated: NODDI metrics came from a nonclinical MRI protocol and ex vivo fixed tissue; fixation can alter tissue microstructure, so values may not fully represent in vivo counterparts.
  • Paper-stated: only four specimens were used, and 2D histological sections (10 μm) were much thinner than MRI slices (800 μm), so not all tissue contributing to MRI signal was sampled.
  • Paper-stated: Palmgren silver staining may leave nonspecific staining, neurofilament antigenicity can be affected by formalin fixation, and only NODDI and DTI metrics were considered; other diffusion or MRI modalities might show stronger histopathological correlations.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1002/acn3.445

Cites

53 of 53 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

25 of 36 cited sources checked · 11 supported · 0 flagged · accuracy not yet measured

  • Premisecitation

    NDI is a surrogate index equivalent to NSF/(1-MSF).

    “This finding is not surprising as NDI is a surrogate index equivalent to NSF/(1–MSF)26, 43 (post hoc analysis confirms a correlation between these two measures).”
    Quote in the paperSubstantial claim1 of 2 sources checked
    • cites [26] Zhang H, Schneider T, Wheeler‐Kingshott CA, Alexander DC. NODDI: practical in vivo neurite orientation dispersion and density imaging of the human brain. NeuroImage 2012; 61: 1000– 1016. 22484410 source
    • cites [43] Jelescu IO, Zurek M, Winters KV, et al. In vivo quantification of demyelination and recovery using compartment‐specific diffusion MRI metrics validated by electron microscopy. NeuroImage 2016; 132: 104– 114. 26876473 PMC4851889 source

      The Introduction describes axonal water fraction (AWF), and Methods §2.4 gives an analogous calculation, tAWF = tAVF/(1 − MVF). Neither passage establishes that NDI is equivalent to NSF/(1 − MSF); searches for that expression and its component terms found no passage addressing the claimed equivalence. Page numbers were unavailable in our copy. (LLM, conf=0.72)

      Experimental — accuracy not yet measured
  • Premisecitation

    Fixation can alter tissue microstructure.

    “Fixation can alter tissue microstructure47, 48; therefore, NODDI metrics as shown here may not be fully representative of their in vivo counterparts.”
    Quote in the paperSubstantial claim1 of 2 sources checked
    • cites [47] Shepherd TM, Thelwall PE, Stanisz GJ, Blackband SJ. Aldehyde fixative solutions alter the water relaxation and diffusion properties of nervous tissue. Magn Reson Med 2009; 62: 26– 34. 19353660 PMC3188415 source
    • cites [48] Birkl C, Langkammer C, Golob‐Schwarzl N, et al. Effects of formalin fixation and temperature on MR relaxation times in the human brain. NMR Biomed 2016; 29: 458– 465. 26835664 source
  • Findingcitation

    DTI is very sensitive to multiple sclerosis pathology, as shown by strong changes of FA, AD, RD, and MD.

    “Both correlation and linear regression analyses confirm once more that DTI is very sensitive to multiple sclerosis pathology, as shown by strong changes of FA, AD, RD, and MD noticeable on simple visual inspection.36”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [36] Klawiter EC, Schmidt RE, Trinkaus K, et al. Radial diffusivity predicts demyelination in ex vivo multiple sclerosis spinal cords. NeuroImage 2011; 55: 1454– 1460. 21238597 PMC3062747 source

      The Results (sections 3.2 and 3.4; page unavailable) report changes in axial and radial diffusivity and ADC/mean diffusivity in MS spinal cords, with radial diffusivity tracking demyelination. But the study measured relative anisotropy rather than FA, found no significant overall relative-anisotropy difference between MS and control tracts, and does not establish that changes in all four named measures were noticeable on simple visual inspection. (LLM, conf=0.87)

      Experimental — accuracy not yet measured
  • Premisecitation

    Neurofilament antigenicity can be negatively affected by formalin fixation.

    “Regarding the NSF maps, neurofilament antigenicity can be negatively affected by formalin fixation,6 which may be at least in part responsible for some of the heterogeneity shown in our neurofilament immunolabeling.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [6] DeLuca G, Ebers G, Esiri M. Axonal loss in multiple sclerosis: a pathological survey of the corticospinal and sensory tracts. Brain 2004; 127: 1009– 1018. 15047586 source

      Materials and methods describes formalin-fixed tissue, and Validation of Palmgren stain to demonstrate axons reports that neurofilament immunostaining demonstrated axons less consistently than the Palmgren stain. The paper does not say that formalin fixation caused this difference or that it reduced neurofilament antigenicity; no page numbers are available in our copy. (LLM, conf=0.74)

      Experimental — accuracy not yet measured
  • Premisecitation

    NODDI can be set up in clinical systems.

    “Also, we provide unequivocal evidence that NODDI dispersion indices are histologically meaningful, demonstrating their clinical viability since NODDI can be set up in clinical systems.26, 27”
    Quote in the paperSubstantial claim1 of 2 sources checked
    • cites [26] Zhang H, Schneider T, Wheeler‐Kingshott CA, Alexander DC. NODDI: practical in vivo neurite orientation dispersion and density imaging of the human brain. NeuroImage 2012; 61: 1000– 1016. 22484410 source
    • cites [27] Grussu F, Schneider T, Zhang H, et al. Neurite orientation dispersion and density imaging of the healthy cervical spinal cord in vivo. NeuroImage 2015; 111: 590– 601. 25652391 source
  • May combine two claims. 38
  • Classified as probably not a substantial claim. 13

  • May combine two claims.

    Premisecitation

    Spinal cord pathology is an important determinant of permanent neurological disability in multiple sclerosis, with cardinal characteristics being inflammatory demyelination, synaptic, neuronal and axonal loss.

    “Spinal cord pathology is an important determinant of permanent neurological disability with cardinal characteristics being inflammatory demyelination, synaptic, neuronal and axonal loss.6, 7, 8, 9, 10, 11”
    Quote in the paperSubstantial claim3 of 3 sources checked
    • cites [6] DeLuca G, Ebers G, Esiri M. Axonal loss in multiple sclerosis: a pathological survey of the corticospinal and sensory tracts. Brain 2004; 127: 1009– 1018. 15047586 source

      The Introduction (page not provided) describes inflammatory demyelination and axonal loss in multiple sclerosis, and says cumulative axonal loss seems likely to contribute to progressive neurological disability. The Discussion (page not provided) describes substantial spinal-cord tract axon loss but notes that limited clinical information prevented relating this study’s findings to established disability scales. The cited passages do not establish synaptic or neuronal loss as cardinal characteristics, or show that spinal-cord pathology determines permanent disability. (LLM, conf=0.88)

      Experimental — accuracy not yet measured
    • cites [7] Calabrese M, Magliozzi R, Ciccarelli O, et al. Exploring the origins of grey matter damage in multiple sclerosis. Nat Rev Neurosci 2015; 16: 147– 158. 25697158 source

      The introduction (p. 2) describes multiple sclerosis pathology as involving inflammation, demyelination, and axonal and neuronal damage. The discussion of spinal-cord cases (p. 8) links inflammation with diffuse axonal loss. But the paper does not establish spinal-cord pathology as an important determinant of permanent disability or identify synaptic loss as a cardinal feature of spinal-cord pathology. (LLM, conf=0.84)

      Experimental — accuracy not yet measured
    • cites [8] Tallantyre E, Bø L, Al‐Rawashdeh O, et al. Greater loss of axons in primary progressive multiple sclerosis plaques compared to secondary progressive disease. Brain 2009; 132: 1190– 1199. 19420101 source

      The Introduction (page not provided) links inflammatory demyelination to relapse disability and describes neuro-axonal pathology as a hypothesized explanation for progressive disability. The study examines cervical spinal cord demyelination, inflammation and axonal loss, but our copy does not establish synaptic or neuronal loss as cardinal characteristics, and it hedges rather than establishes the cause of permanent disability. (LLM, conf=0.86)

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingcitation

    In control spinal cord specimens, DTI metrics show higher diffusion anisotropy in white matter compared with gray matter.

    “DTI metrics show known trends,36 such as higher diffusion anisotropy in white as compared with gray matter.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [36] Klawiter EC, Schmidt RE, Trinkaus K, et al. Radial diffusivity predicts demyelination in ex vivo multiple sclerosis spinal cords. NeuroImage 2011; 55: 1454– 1460. 21238597 PMC3062747 source

      The Results (§3.2) compare DTI measurements in white-matter tracts from control and MS spinal cords, not white matter with gray matter in controls. A search for gray matter found only its use as an anatomical boundary in Methods (§2.5); our copy does not report the claimed anisotropy comparison. (LLM, conf=0.76)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Findingcitation

    DTI metrics highlight focal pathology with increased diffusivity and changes in FA.

    “DTI metrics highlight focal pathology as well, with increased diffusivity and changes in FA, similarly to what has been previously reported.36”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [36] Klawiter EC, Schmidt RE, Trinkaus K, et al. Radial diffusivity predicts demyelination in ex vivo multiple sclerosis spinal cords. NeuroImage 2011; 55: 1454– 1460. 21238597 PMC3062747 source

      The Results (§3.4 Myelin analysis; page unavailable) show increased diffusivity in demyelinated regions and a change in relative anisotropy (RA), supporting the focal-pathology point. The study did not report a corresponding FA result of its own; FA findings appear in its discussion of other studies (§4 Discussion). (LLM, conf=0.86)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Findingcitation

    In controls, NODDI NDI and ODI show contrast between gray and white matter: NDI is higher in white matter than gray matter, while ODI is higher in gray than white matter.

    “In controls, NODDI NDI and ODI show contrast between gray and white matter, in line with recent in vivo results.27 NDI is higher in white matter than in gray matter, while the opposite holds for ODI.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [27] Grussu F, Schneider T, Zhang H, et al. Neurite orientation dispersion and density imaging of the healthy cervical spinal cord in vivo. NeuroImage 2015; 111: 590– 601. 25652391 source

  • May combine two claims.

    Study design and conductcitation

    The data set of 48 data points suffices to fit the regression models and evaluate correlations.

    “Practically, this led to the construction of a data set of 48 data points, whose size suffices to fit the regression models and evaluate correlations (other high‐impact studies on MRI‐histopathology correlation relied on the same amount of data20 or even less17).”
    Quote in the paperSubstantial claim2 of 2 sources checked
    • cites [20] Moll NM, Rietsch AM, Thomas S, et al. Multiple sclerosis normal‐appearing white matter: Pathology–imaging correlations. Ann Neurol 2011; 70: 764– 773. 22162059 PMC3241216 source
    • cites [17] Wisnieff C, Ramanan S, Olesik J, et al. Quantitative susceptibility mapping (QSM) of white matter multiple sclerosis lesions: interpreting positive susceptibility and the presence of iron. Magn Reson Med 2015; 74: 564– 570. 25137340 PMC4333139 source

      The cited paper’s Methods and Results describe MRI–histopathology comparisons, but our copy does not report a dataset of 48 data points or establish that 48 suffice for regression and correlation analysis. The Discussion instead says that differing slice thicknesses make meaningful voxel-based linear regression difficult; it does not assess the sufficiency of a 48-point dataset (searched: 48 data points, regression, correlations). (LLM, conf=0.76)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Interpretationcitation

    The results agree well with preliminary findings in vivo, providing confidence that the conclusions will most likely hold true in clinical scenarios.

    “Nevertheless, our results agree well with preliminary findings in vivo, 39, 49 providing confidence that the conclusions of this study will most likely hold true in clinical scenarios.”
    Quote in the paperSubstantial claim1 of 2 sources checked
    • cites [39] Schneider T, Brownlee W, Zhang H, et al. Application of multi‐shell NODDI in multiple sclerosis. Proceedings of the 22nd scientific meeting of the International Society for Magnetic Resonance in Medicine. 2014; 0019.

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [49] Schneider T, Brownlee W, Zhang H, et al. Sensitivity of multi‐shell NODDI to multiple sclerosis white matter changes: a pilot study. Funct Neurol 2017; 32: 97– 101. 28676143 PMC5507159 source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductcitation

    Areas of nonspecific staining were excluded with a combination of manual and automatic segmentation to calculate CV.

    “excluded areas of nonspecific staining with a combination of manual and automatic segmentation to calculate CV, given their potential to bias this metric.33”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [33] Grussu F, Schneider T, Yates RL, et al. A framework for optimal whole‐sample histological quantification of neurite orientation dispersion in the human spinal cord. J Neurosci Methods 2016; 273: 20– 32. 27497747 source

      Section 2.2.3 says the authors manually masked areas with large vessels and concentrations of neuronal cell bodies. Sections 2.3 and 4.7 describe automated neurite segmentation used to assess CV while excluding non-neurite pixels, but section 4.7 presents automatic removal of nonspecifically stained areas as a future improvement—not something the study did. (LLM, conf=0.88)

      Experimental — accuracy not yet measured

  • Classified as probably not a substantial claim.

    Premisecitation

    Neurite morphology is a substrate of neural function.

    “Our study tests the hypothesis that the variability of neurite orientations is a biomarker of multiple sclerosis pathology, and is motivated by recent findings demonstrating that neurite morphology is a substrate of neural function23, 24 and is affected by multiple sclerosis.25”
    Quote in the paperSubstantial claim2 of 2 sources checked
    • cites [23] Nazeri A, Chakravarty MM, Rotenberg DJ, et al. Functional consequences of neurite orientation dispersion and density in humans across the adult lifespan. J Neurosci 2015; 35: 1753– 1762. 25632148 PMC4308611 source
    • cites [24] Dickstein DL, Kabaso D, Rocher AB, et al. Changes in the structural complexity of the aged brain. Aging Cell 2007; 6: 275– 284. 17465981 PMC2441530 source

  • Classified as probably not a substantial claim.

    Premisecitation

    Neurite morphology is affected by multiple sclerosis.

    “Our study tests the hypothesis that the variability of neurite orientations is a biomarker of multiple sclerosis pathology, and is motivated by recent findings demonstrating that neurite morphology is a substrate of neural function23, 24 and is affected by multiple sclerosis.25”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [25] Jürgens T, Jafari M, Kreutzfeldt M, et al. Reconstruction of single cortical projection neurons reveals primary spine loss in multiple sclerosis. Brain 2016; 139: 39– 46. 26667278 source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingcitation

    Known contrasts are observed between controls' gray and white matter for DTI metrics.

    “Regarding DTI metrics, known contrasts are observed between controls’ gray and white matter.27, 36”
    Quote in the paperNot a substantial claim2 of 2 sources checked
    • cites [27] Grussu F, Schneider T, Zhang H, et al. Neurite orientation dispersion and density imaging of the healthy cervical spinal cord in vivo. NeuroImage 2015; 111: 590– 601. 25652391 source
    • cites [36] Klawiter EC, Schmidt RE, Trinkaus K, et al. Radial diffusivity predicts demyelination in ex vivo multiple sclerosis spinal cords. NeuroImage 2011; 55: 1454– 1460. 21238597 PMC3062747 source

      Methods §2.3 describes control DTI measurements in white-matter tracts, and Methods §2.5 mentions gray–white matter boundaries as registration landmarks; neither compares DTI metrics between the two tissues. Results §3.2 compares white-matter tracts in controls and MS subjects, not gray and white matter within controls. A search for control gray–white matter DTI contrasts found no passage addressing the claim; page numbers were unavailable. (LLM, conf=0.74)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Findingown data

    In nonlesional areas of the upper thoracic multiple sclerosis specimen, gray matter compared with white matter shows increased CV, increased ODI, and decreased NDI.

    “In nonlesional areas, differences between gray and white matter are observed with increased CV (histology), increased ODI (MRI), and decreased NDI (MRI) in gray compared to white matter as seen in controls.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingreasoning

    NDI always drops dramatically within lesions, similar to myelin density MSF.

    “In contrast, NDI always drops dramatically within lesions, similar to myelin density MSF.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    In the upper thoracic multiple sclerosis cord specimen, there is a solitary chronic inactive demyelinated white matter lesion with paucity of microglial inflammation, relative axonal preservation, and associated astrogliosis.

    “In the upper thoracic cord specimen, we observe a solitary chronic inactive demyelinated white matter lesion with a paucity of microglial inflammation, relative axonal preservation, and associated astrogliosis.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    In the upper lumbar multiple sclerosis lesions, compared with nonlesional areas, there is significant reduction in neurofilament density and absence of pronounced astrogliosis.

    “the gray and white lesions in the lumbar cord specimen show a significant reduction in neurofilament density (reflecting axonal loss) and an absence of pronounced astrogliosis.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    In the upper lumbar multiple sclerosis cord specimen, separate demyelinated lesions in gray and white matter are seen; all are chronic inactive with loss of myelin and absence of microglial inflammation.

    “In the upper lumbar cord specimen, separate demyelinated lesions in each of gray and white matter, are seen. All lesions are chronic inactive demonstrating loss of myelin and an absence of microglial inflammation.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    In nonlesional tissue, NODDI metrics in gray and white matter follow a similar pattern as that described in controls, as do DTI metrics.

    “In nonlesional tissue, NODDI metrics in gray and white matter follow a similar pattern as that described in controls, as well as those from DTI.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    In the upper thoracic multiple sclerosis lesion, compared with nonlesional white matter, MSF is decreased, μGSF is decreased, ASF is increased, and NSF is unchanged.

    “MSF is decreased (myelin loss), μGSF is decreased (microglia reduction), ASF is increased (astrogliosis), and NSF is unchanged (neurofilament preservation) in the focal white matter lesion compared to nonlesional white matter areas.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingcitation

    Lesional matter shows hypointense NDI and ODI maps and hyperintense IVF relative to nonlesional areas.

    “In contrast, lesional matter shows hypointense NDI and ODI maps and hyperintense IVF relative to nonlesional areas, in line with recent findings in vivo.39”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [39] Schneider T, Brownlee W, Zhang H, et al. Application of multi‐shell NODDI in multiple sclerosis. Proceedings of the 22nd scientific meeting of the International Society for Magnetic Resonance in Medicine. 2014; 0019.

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Findingown data

    In the upper thoracic multiple sclerosis lesion, NODDI MRI maps demonstrate increased IVF and decreased NDI and ODI compared to nonlesional areas.

    “NODDI MRI maps demonstrate increased IVF and decreased NDI and ODI in this lesion compared to nonlesional areas.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    In the upper lumbar multiple sclerosis lesions, DTI maps show decreased FA and increased AD, RD, and MD in lesional areas compared to nonlesional tissue.

    “DTI maps show changes in FA (decreased) and AD, RD, and MD (all increased) in lesional areas, as compared to nonlesional tissue.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    In the upper thoracic multiple sclerosis lesion, DTI metrics show variations in FA (both increase and decrease) and increased AD, RD, and MD.

    “DTI metrics show variations in FA (with both areas of increase and of decrease), and increased AD, RD, and MD.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    In control spinal cord specimens, NDI is higher in white matter than gray matter, whereas ODI is higher in gray matter than white matter.

    “In contrast, NDI is higher in white compared to gray matter whereas the converse is true for ODI.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    In the upper lumbar multiple sclerosis lesions, compared with nonlesional areas, MSF, NSF, and μGSF are decreased and ASF is unchanged.

    “a decrease in MSF (myelin loss), NSF (axonal loss), and μGSF (microglia reduction) and no change in ASF (astrogliosis) in lesional compared to nonlesional areas”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    In post mortem multiple sclerosis spinal cord specimens, both NODDI ODI and histological CV show a trend toward lower neurite orientation dispersion in demyelinated lesions compared with nonlesional tissue.

    “We report from both NODDI and histology a trend toward lower neurite dispersion in demyelinated lesions”
    • Also stated in: § Results · plain
      “Both MRI (ODI) and histology (CV) reveal a trend of lower neurite orientation dispersion in multiple sclerosis demyelinated lesions”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    In controls, linear regression model 2 shows NDI and ODI have negative and positive association with CV respectively.

    “Model number 2: in controls, NDI and ODI show, respectively a negative and positive association with CV (NDI: β = −0.789, P = 0.0002; ODI: β = 0.838, P < 10−5)”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    In multiple sclerosis cases (model 4), AD, RD and MD are all negatively associated to MSF and NSF.

    “AD, RD and MD are all negatively associated to MSF (AD: β = −0.579; P = 0.0001; RD: β = −0.447; P = 0.009; MD: β = − 0.494; P = 0.002) and NSF (AD: β = −0.384; P = 0.006; RD: β = −0.378, P = 0.025; MD: β = − 0.396; P = 0.012).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    For model 4, very low adjusted coefficients of determination are observed for ODI and IVF and NDI in control cases, while a higher value is obtained for NDI in multiple sclerosis.

    “For model 4, very low values of this coefficient are observed for ODI and IVF (control cases: −0.010 for ODI, 0.194 for IVF; multiple sclerosis cases: 0.137 for ODI, 0.008 for IVF) and NDI in control cases (−0.118), while a higher value is obtained for NDI in multiple sclerosis (0.577).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    In multiple sclerosis cases (model 5), NDI is positively associated with μGSF (β = 0.508; P = 0.009), while AD, RD and MD are negatively associated to μGSF.

    “In multiple sclerosis cases, NDI is positively associated with μGSF (β = 0.508; P = 0.009), while AD (β = −0.581; P = 0.002), RD (β = −0.527; P = 0.006) and MD (β = −0.558; P = 0.003) are negatively associated to μGSF.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Model 5: no significant associations between NODDI metrics and glial fractions ASF and μGSF are detected in controls, while DTI AD and MD show a positive association with μGSF.

    “Model number 5: no significant associations between NODDI metrics and glial fractions ASF and μGSF are detected in controls, while DTI AD (β = 0.632; P = 0.004) and DTI MD (β = 0.501; P = 0.030) show a positive association with μGSF.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Linear regression models confirm that NODDI NDI offers sensitivity to the density of neuronal elements (positive association with NDI-NSF in model 3), but is also strongly influenced by the local amount of myelin.

    “Finally, linear regression models confirm that NODDI NDI offers sensitivity to the density of neuronal elements (positive association with NDI‐NSF in model 3), but is also strongly influenced by the local amount of myelin.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    In multiple sclerosis cases (model 4), NDI is positively associated with both MSF (β = 0.610; P = 0.0002) and NSF (β = 0.288; P = 0.050).

    “In multiple sclerosis cases, NDI is positively associated with both MSF (β = 0.610; P = 0.0002) and NSF (β = 0.288; P = 0.050).”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    MRI metrics may be able to detect pathology that extends from focal lesions in the peri-plaque area.

    “For instance, a preliminary investigation reported as supplementary material (Data S4) highlights how MRI metrics may be able to detect pathology that extends from focal lesions in the peri‐plaque area.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingreasoning

    In multiple sclerosis, ODI shows only a weak correlation with MSF, not confirmed by linear regression.

    “ODI shows only a weak correlation with MSF in multiple sclerosis cases (not confirmed by linear regression), likely due to the similar contrast of the two between lesional and nonlesional tissue.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisecitation

    Conventional MRI of the multiple sclerosis spinal cord has low specificity for underlying pathological processes, and radiological changes fail to consistently correlate with measures of disability.

    “Conventional magnetic resonance imaging (MRI) of the multiple sclerosis spinal cord is limited by low specificity regarding the underlying pathological processes”
    • Also stated in: § Introduction · plain
      “However, the pathological substrate of radiological changes are relatively nonspecific and fail to consistently correlate with measures of disability.12”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [12] Kearney H, Miller DH, Ciccarelli O. Spinal cord MRI in multiple sclerosis ‐ diagnostic, prognostic and clinical value. Nat Rev Neurol 2015; 11: 327– 338. 26009002 source

  • May combine two claims.

    Interpretationown data

    NODDI is sensitive and specific to the underlying orientation dispersion as measured by histology in post mortem spinal cord tissue from multiple sclerosis and nonneurological controls (significant association between MRI and histological dispersion: P < 0.001).

    “NODDI is sensitive and specific to the underlying orientation dispersion as measured by histology in post mortem spinal cord tissue from multiple sclerosis and nonneurological controls (significant association between MRI and histological dispersion: P < 0.001).”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S4 (based). May combine two claims.

    Interpretationnone given

    NODDI metrics are specific to neurons and are not influenced by features of the extra-neuronal space, such as density of glial cells.

    “Furthermore, NODDI metrics are specific to neurons and are not influenced by features of the extra‐neuronal space, such as density of glial cells.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationown data

    The complexity of neurite composition can be altered not only in the multiple sclerosis cortex, but also in the functionally relevant spinal cord.

    “Our work highlights how the complexity of neurite composition can be altered not only in the multiple sclerosis cortex, but also in the functionally relevant spinal cord.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisecitation

    When myelin is lost, the local T2 increases, effectively increasing the amount of MRI-visible water and causing NDI to decrease.

    “When myelin is lost, the local T2 increases,46 effectively increasing the amount of MRI‐visible water and causing NDI to decrease.26”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites [46] Stevenson V, Parker G, Barker G, et al. Variations in T1 and T2 relaxation times of normal appearing white matter and lesions in multiple sclerosis. J Neurol Sci 2000; 178: 81– 87. 11018698 source
    • cites [26] Zhang H, Schneider T, Wheeler‐Kingshott CA, Alexander DC. NODDI: practical in vivo neurite orientation dispersion and density imaging of the human brain. NeuroImage 2012; 61: 1000– 1016. 22484410 source

  • Classified as probably not a substantial claim.

    Interpretationnone given

    NODDI NDI is strongly influenced by variations of myelination, limiting its interpretability without myelin mapping techniques.

    “Importantly, NODDI NDI offers sensitivity to the local density of axon/dendrites but is also strongly influenced by variations of myelination, limiting its interpretability without the support of myelin mapping techniques.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationown data

    DTI metrics are less specific and influenced by several biophysical substrates.

    “while DTI metrics are less specific and influenced by several biophysical substrates.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationreasoning

    The in vivo application of neurite orientation dispersion mapping has the potential to cast light onto pathophysiological processes relevant to multiple sclerosis and to provide a clinically sensitive outcome measure in clinical trial and practice settings for prognosis and monitoring treatment response.

    “The in vivo application of neurite orientation dispersion mapping has the potential to not only cast light onto pathophysiological processes relevant to multiple sclerosis, but also to provide a clinically sensitive outcome measure in clinical trial and practice settings for prognosis and monitoring treatment response.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductnone given

    NODDI metrics were obtained with a nonclinical MRI protocol and from ex vivo fixed tissue.

    “NODDI metrics were obtained with a nonclinical MRI protocol and from ex vivo fixed tissue.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationown data

    NODDI provides a histologically meaningful measure of neurite dispersion that can be used to unravel previously undetected layers of complexity of multiple sclerosis pathology in the spinal cord.

    “Moreover, our work highlights how a clinically viable quantitative MRI method, NODDI, provides a histologically meaningful measure of neurite dispersion that can be used to unravel previously undetected layers of complexity of multiple sclerosis pathology in the spinal cord.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Study design and conductreasoning

    Two different histological sections per MRI slice (200 μm apart) were derived and averaged for analysis to account for thickness mismatch.

    “To account for this mismatch, we derived two different histological sections per MRI slice (200 μm apart), and averaged the histological indices for analysis.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    The number of specimens employed in this study (four) is limited.

    “Furthermore, we also acknowledge that the number of specimens employed in this study (i.e., four) is limited, although results are compelling and exciting.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Interpretationreasoning

    The paper reports a heretofore undescribed layer of complexity of multiple sclerosis pathology.

    “In so doing, we report a heretofore undescribed layer of complexity of multiple sclerosis pathology.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.

    Interpretationnone given

    NODDI indices are designed to measure geometrical features of neurite morphology, but in practice they can be influenced by other factors, such as myelin for NDI.

    “We stress that NODDI metrics should always be interpreted with care: NODDI indices are designed to measure geometrical features of neurite morphology, but in practice they can be influenced by other factors, as for example myelin for the case of NDI.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Study design and conductreasoning

    Not all tissue that contributed to the MRI signal was sampled, possibly explaining some differences between MRI and histology.

    “Nonetheless, not all the tissue that contributed to the MRI signal was sampled, possibly explaining some of the differences between MRI and histology.”
    Quote in the paperNot a substantial claim
48 of 48

  • The paper relies on this without support.

    The precise nature of how multiple sclerosis spinal cord pathological alterations impact clinical phenotypes remains unknown.

    “The precise nature of how these pathological alterations impact clinical phenotypes remains unknown.”

  • The paper relies on this without support.

    Neuropathological analyses capture the micro-architectural substrates of the MR signal.

    “Histological indices were considered as the predictors of the MRI metrics, as neuropathological analyses capture the micro‐architectural substrates of the MR signal.”

  • The paper relies on this without support.

    The four post mortem spinal cord specimens are representative of multiple sclerosis spinal cord pathology.

    “Furthermore, we also acknowledge that the number of specimens employed in this study (i.e., four) is limited, although results are compelling and exciting.”
    “Therefore, further research is required to confirm the changes in neurite dispersion due to multiple sclerosis pathology reported here.”
3 of 3

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • ODI decreases are thought to indicate loss of crossing fibers and decreased dendritic arborizations evident in various central nervous system disorders.

    “ODI decreases are thought to indicate loss of crossing fibers 50,51 and decreased dendritic arborizations evident in various central nervous system disorders.45,47”

    From White Matter Changes after Neurobehavioral Therapy for Functional Seizures (2023) · cites this paper as [47]

  • ODI has shown a stronger correlation than FA with orientation dispersion measured by histology.

    “ODI has shown a stronger correlation than has FA with orientation dispersion measured by histology (Grussu et al., 2017;Schilling et al., 2018), indicating that NODDI has the ability, at least partly, to disentangle the effect of orientation dispersion.”

    From NODDI in clinical research (2020) · cites this paper as Grussu et al., 2017

  • In postmortem human spinal cord, decreased ODI in MS lesions and excellent agreement between ODI and its histological counterpart were observed.

    “In postmortem human spinal cord, (Grussu et al., 2017) observed decreased ODI in MS lesions and an excellent agreement between ODI and its histological counterpart.”

    From NODDI in clinical research (2020) · cites this paper as Grussu et al., 2017

  • NDI should be interpreted in relation to both neurite density and myelin density.

    “The study of postmortem spinal cord by (Grussu et al., 2017) also revealed that NDI should be interpreted in relation to both neurite density and myelin density.”

    From NODDI in clinical research (2020) · cites this paper as Grussu et al., 2017

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • Claim of no glial influence conflicts with NDI-microglia association

    Minor

    The discussion says NODDI metrics are not influenced by glial-cell density, whereas the reported model 5 finds a significant positive association between NDI and microglia staining fraction in multiple sclerosis cases.

  • Opposite gray-white matter CV pattern attributed to controls

    Minor

    The lumbar multiple sclerosis results say CV decreases in gray matter relative to white matter and describe this as the pattern seen in controls. The control results state that CV is higher in gray matter.

2 of 2

Cites

Flagged first, then by citations

1 of the 54 references couldn’t be matched to a known paper.

Research works printed but not linked: 1

  1. SchneiderT BrownleeW ZhangH et al.Application of multi‐shell NODDI in multiple sclerosis. Proceedings of the 22nd scientific meeting of the International Society for Magnetic Resonance in Medicine.2014;0019.

Cited by

Sorted by citations

Related

Authors
Grussu, F., Schneider, T., Tur, C., Yates, R. L., Tachrount, M., Ianuş, A., Yiannakas, M. C., Newcombe, J., Zhang, H., Alexander, D. C., DeLuca, G. C., & Gandini Wheeler‐Kingshott, C. A. M.
Affiliations
  • National Hospital for Neurology and Neurosurgery (GB)
  • Philips (United Kingdom) (GB)
  • University of Oxford (GB)
  • UCL Queen Square Institute of Neurology (GB)
  • Centre for Medical Image Computing Department of Computer Science University College London London United Kingdom
  • University College London (GB)
  • NMR Research Unit Department of Neuroinflammation Queen Square MS Centre UCL Institute of Neurology University College London London United Kingdom
Corresponding author
DeLuca, Gabriele C.
ORCID
Venue
Annals of Clinical and Translational Neurology
Publisher
Wiley
Published
2017-08-15
Volume
4
Issue
9
Pages
663-679
ISSN
2328-9503 (ISSN-L)
DOI
10.1002/acn3.445
PMC
PMC5590517
OpenAlex
W2747366609
Cited by
291
References
54
Citations by year
2026: 17 · 2025: 24 · 2024: 36 · 2023: 35 · 2022: 34 · 2021: 48 · 2020: 31 · 2019: 31 · 2018: 30 · 2017: 4
FWCI
18.17
Type
journal article
Language
en
Open access
Licence
Locations
  • Annals of Clinical and Translational Neurology (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • PubMed Central (submittedVersion, cc-by)
  • UCL Discovery (University College London) (submittedVersion)
Funders
  • University College London
  • Horizon 2020 Framework Programme ((H2020)‐EU.3.1 CDS‐QUAMRI (634541))
  • Engineering and Physical Sciences Research Council (EP/M020533/1, M507970, G007748, I027084, M020533, N018702, H2020‐EU.3.1 (634541, 666992‐2))
  • Multiple Sclerosis Society (892/08)
  • National Institute for Health Research
  • European Committee for Treatment and Research in Multiple Sclerosis
Topics
  • Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › Advanced Neuroimaging Techniques and Applications
  • Health Sciences › Medicine › Pathology and Forensic Medicine › Multiple Sclerosis Research Studies
  • Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › Advanced MRI Techniques and Applications
Keywords
multiple sclerosis, myelopathy, neurite orientation dispersion, NODDI, demyelination, diffusion MRI, diffusion tensor imaging, histopathological confirmation, neurofilaments, myelin, astrocytes, microglia, biomarkers

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-13.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-13.

Cited sources are checked by gpt-6-sol.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review