Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Comment
  • Published:

Advancing neuroprognostication in neonatal encephalopathy: promise and limitations of diffusion kurtosis imaging

Pediatric Research (2025)Cite this article

Accurate neuroprognostication in infants with neonatal encephalopathy (NE) remains a pressing clinical challenge.1,2,3 In this volume of Pediatric Research, Moss et al.4 suggest that using diffusion kurtosis imaging (DKI) instead of the more commonly used diffusion tensor imaging (DTI) would improve neuroprognostication in NE due to presumed hypoxic-ischemic encephalopathy (HIE). The authors provide a wide array of compelling arguments, and a demonstration, albeit in a small sample, of the increased sensitivity to white matter injury that DKI offers relative to DTI. Moss et al.4 point out that DTI assumes a Gaussian probability distribution function for water diffusion dynamics, but that non-Gaussian diffusion occurs in regions with crossing fibers, at extra- and intra-cellular membrane barriers, and where the tissue architecture is complex in some other way, including due to injury. They argue that assuming water diffusion in the brain follows a Gaussian model limits DTI’s ability to accurately detect microstructural abnormalities in the developing, largely unmyelinated brain after injury. The investigators point out that the non-Gaussian diffusion becomes more obvious at higher diffusion weightings, which provides the basis of DKI. They also argue that studies have shown that DKI metrics, such as mean kurtosis (MK), are better at detecting and characterizing early and subtle brain injury after adult acute ischemic stroke. Further, they argue that kurtosis metrics improve predictions of white matter recovery, and long-term outcome compared with diffusivity metrics. They suggest that, though studies of neonatal brain injury have until now primarily used DTI, DKI may offer advantages in the developing brain similar to those seen in adult stroke.5,6,7,8 They present a small demonstration that there are such advantages to the use of DKI instead of DTI for neuroprognostication in NE. Their design included a cohort receiving only therapeutic hypothermia (TH) and scanned shortly after rewarming; and another treated with N-acetylcysteine and vitamin D, in addition to TH, and scanned shortly after rewarming and again weeks later so that the impact of the treatment could be assessed. Notably, the timing of brain MRI differed between cohorts, and these timing differences could influence diffusion metrics, as imaging earlier in the subacute/recovery period may better capture the evolution of injury. In each case, the cohorts consisted of infants with moderate to severe NE, and the analyses assessed the extent to which DKI better distinguished the moderate cases from the severe cases than did DTI. The study, though using a sample size too small to allow correction for multiple comparisons, and involving different scanner models and scan protocols for each cohort, and lacking a normative control group, nonetheless seemed to show that DKI better distinguished the severe cases from the moderate cases, particularly DKI based on the recovery-phase scans, and was more sensitive to the infants’ response to therapeutic intervention.

Recently, advanced neuroimaging techniques have gained attention as promising tools to improve neuroprognostication in NE.3,9,10 Overall, Moss et al.4 present evidence that DKI may offer valuable prognostic insight in NE. Specifically, diffusion metrics such as MK and kurtosis fractional anisotropy (KFA) appeared to align with both the clinical severity of white matter injury and neurodevelopmental outcomes. Notably, the recovery-phase scans yielded stronger associations with gross motor and cognitive scores compared to standard diffusivity measures. Additionally, DKI parameters were sensitive to treatment effects, particularly to N-acetylcysteine and vitamin D, highlighting the potential of DKI not only in characterizing hypoxic-ischemic injury but also in tracking therapeutic response. In our opinion, the study’s strengths are clear: it samples two clinically meaningful windows and uses a targeted region of interest strategy focused on tracts that are vulnerable to hypoxic-ischemic injury, such as the posterior limb of the internal capsule, corticospinal tracts at the cerebral peduncle, corpus callosum, and posterior thalamic radiations. This aligns with expected motor and cognitive phenotypes and improves interpretability over whole-brain voxelwise assessments. Notably, the authors also collected data on the Area Deprivation Index to account for environmental modifiers of neurodevelopment, which is uncommon but potentially important in neonatal neuroimaging studies. Operationally, scans were obtained unsedated after rewarming, and DKI is positioned as a modest add-on to standard protocols (roughly between 5 and 8 min), enabling feasibility and real-world adoption.

This is a preview of subscription content, access via your institution

Access options

Buy this article

  • Purchase on SpringerLink
  • Instant access to the full article PDF.

USD 39.95

Prices may be subject to local taxes which are calculated during checkout

References

  1. Cizmeci, M. N., van Steenis, A., Christensen, R., Steggerda, S. J. & de Vries, L. S. Leveraging neonatal neuroimaging for neuroprognostication in presumed hypoxic-ischemic encephalopathy: a framework for clinicians. Semin. Perinatol. 152126 https://doi.org/10.1016/j.semperi.2025.152126 (2025).

  2. Cizmeci, M. N., Christensen, R., van Steenis, A. & de Vries, L. S. Neuroprognostication in neonatal encephalopathy due to presumed hypoxic-ischemic encephalopathy. Pediatr Res. https://doi.org/10.1038/s41390-025-04058-1 (2025).

  3. Chawla, V. et al. Emerging modalities for neuroprognostication in neonatal encephalopathy: harnessing the potential of artificial intelligence. Pediatr. Res. https://doi.org/10.1038/s41390-025-04336-y (2025).

  4. Moss, H. G. et al. Neonatal encephalopathy: a diffusional kurtosis imaging analysis of white matter to assess injury severity and recovery. Pediatr Res. https://doi.org/10.1038/s41390-025-04434-x (2025).

  5. Yu, X. et al. Changes in the Corticospinal tract beyond the ischemic lesion following acute hemispheric stroke: a diffusion kurtosis imaging study. J. Magn. Reson Imaging 52, 512–519 (2020).

    Article  PubMed  Google Scholar 

  6. Hui, E. S. et al. Stroke assessment with diffusional Kurtosis Imaging. Stroke 43, 2968–2973 (2012).

    Article  PubMed  PubMed Central  Google Scholar 

  7. Jensen, J. H. et al. Preliminary observations of increased diffusional kurtosis in human brain following recent cerebral infarction. NMR Biomed. 24, 452–457 (2011).

    Article  PubMed  Google Scholar 

  8. Spampinato, M. V. et al. Diffusional Kurtosis imaging and motor outcome in acute ischemic stroke. AJNR Am. J. Neuroradiol. 38, 1328–1334 (2017).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  9. Lewis, J. D. et al. Automated Neuroprognostication Via Machine Learning in Neonates with Hypoxic-Ischemic Encephalopathy. Ann. Neurol. 97, 791–802 (2025).

    Article  PubMed  Google Scholar 

  10. Onda, K. et al. Quantification of Diffusion Magnetic Resonance Imaging for Prognostic Prediction of Neonatal Hypoxic-Ischemic Encephalopathy. Dev. Neurosci. 46, 55–68 (2024).

    Article  CAS  PubMed  Google Scholar 

  11. McJury, M. & Shellock, F. G. Auditory Noise Associated with Mr Procedures: A Review. J. Magn. Reson Imaging 12, 37–45 (2000).

    Article  CAS  PubMed  Google Scholar 

  12. Foster, J. R., Hall, D. A., Summerfield, A. Q., Palmer, A. R. & Bowtell, R. W. Sound-Level Measurements and Calculations of Safe Noise Dosage During EPI at 3 T. J. Magn. Reson. Imaging 12, 157–163 (2000).

    Article  CAS  PubMed  Google Scholar 

  13. Merisaari, H. et al. Effect of Number of Diffusion Encoding Directions in Neonatal Diffusion Tensor Imaging Using Tract-Based Spatial Statistical Analysis. Eur. J. Neurosci. 58, 3827–3837 (2023).

    Article  CAS  PubMed  Google Scholar 

  14. Woodward, K., Spencer, A. P. C., Jary, S. & Chakkarapani, E. Factors Associated with Mri Success in Children Cooled for Neonatal Encephalopathy and Controls. Pediatr. Res 93, 1017–1023 (2023).

    Article  PubMed  Google Scholar 

  15. Palombo, M. et al. Esr Essentials: Diffusion-Weighted MRI-Practice Recommendations by the European Society for Magnetic Resonance in Medicine and Biology. Eur. Radiol. (2025).

  16. Seider, N. A. et al. Accuracy and Reliability of Diffusion Imaging Models. Neuroimage 254, 119138 (2022).

    Article  PubMed  Google Scholar 

  17. Henriques, R. N., Jespersen, S. N., Jones, D. K. & Veraart, J. Toward More Robust and Reproducible Diffusion Kurtosis Imaging. Magn. Reson. Med. 86, 1600–1613 (2021).

    Article  PubMed  PubMed Central  Google Scholar 

  18. Marrale, M. et al. Physics, Techniques and Review of Neuroradiological Applications of Diffusion Kurtosis Imaging (DKI). Clin. Neuroradiol. 26, 391–403 (2016).

    Article  CAS  PubMed  Google Scholar 

Download references

Funding

No internal or external funding was secured for this article.

Author information

Authors and Affiliations

  1. Division of Neonatology, The Hospital for Sick Children, and Department of Paediatrics, University of Toronto, Toronto, ON, Canada

    John D. Lewis & Mehmet N. Cizmeci

  2. Division of Newborn Medicine, Department of Pediatrics, Boston Children’s Hospital and Harvard Medical School, Boston, MA, USA

    Brian T. Kalish

Authors
  1. John D. Lewis
    View author publications

    Search author on:PubMed Google Scholar

  2. Brian T. Kalish
    View author publications

    Search author on:PubMed Google Scholar

  3. Mehmet N. Cizmeci
    View author publications

    Search author on:PubMed Google Scholar

Contributions

J.D.L. conceptualized and designed the article, interpreted the data, drafted the article and approved the final version to be published. B.T.K. conceptualized and designed the article, interpreted the data, and approved the final version to be published. MNC drafted and critically reviewed the article and approved the final version to be published.

Corresponding author

Correspondence to Mehmet N. Cizmeci.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Lewis, J.D., Kalish, B.T. & Cizmeci, M.N. Advancing neuroprognostication in neonatal encephalopathy: promise and limitations of diffusion kurtosis imaging. Pediatr Res (2025). https://doi.org/10.1038/s41390-025-04714-6

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Version of record:

  • DOI: https://doi.org/10.1038/s41390-025-04714-6

Search

Advanced search

Quick links