Diagnostic value of diffusion kurtosis imaging for brain injury in term neonates with hypoxic-ischemic encephalopathy.

Background: Neonatal hypoxic-ischemic encephalopathy (HIE) stands as a primary cause of neonatal brain injury, mortality, and long-term neurodevelopmental impairment, imposing a considerable burden on public health systems worldwide. Despite advances in clinical practice, accurate early diagnosis and reliable severity stratification of HIE remain formidable challenges when relying solely on conventional neuroimaging modalities. Materials and methods: A cohort of 66 term neonates with clinically confirmed HIE and 25 age-matched healthy controls underwent cranial magnetic resonance imaging (MRI) combined with diffusion kurtosis imaging (DKI) within the first 28 days of life. A region-of-interest (ROI)-based quantitative analysis was conducted to quantify kurtosis and diffusion metrics across distinct white matter and deep gray matter regions. The diagnostic utility and severity-stratifying capacity of these DKI parameters were evaluated using receiver operating characteristic (ROC) curve analysis, while their potential correlations with 1-min and 5-min Apgar scores were further explored. Results: ROI-based analysis revealed significant differences in DKI parameters between neonates with HIE and healthy controls across multiple brain regions. Notably, DKI parameters derived from deep gray matter exhibited superior diagnostic performance and severity stratification capacity, with several kurtosis indices achieving an area under the ROC curve (AUC) greater than 0.90. Subgroup comparisons further demonstrated that specific DKI parameters differed significantly between mild and moderate, and between mild and moderate-to-severe HIE cases. Among all measured indices, the mean kurtosis of the corpus callosum yielded relatively high discriminative efficacy. Additionally, a subset of DKI parameters showed significant correlations with Apgar scores, with a greater number of parameters correlating strongly with the 5-min Apgar score. Conclusion: DKI yields robust quantitative biomarkers that facilitate the accurate diagnosis and reliable severity stratification of neonatal HIE, with deep gray matter emerging as a pivotal indicator of brain injury severity. These findings underscore the potential value of advanced diffusion-weighted imaging techniques in enabling early risk stratification and standardized assessment of neonatal brain injury, which carries profound implications for the optimization of public health-oriented neonatal care strategies.