Germinal matrix-intraventricular hemorrhage (GMH-IVH) poses a significant challenge in extremely preterm infants despite advancements in neonatal care. While survival rates for extremely preterm infants have increased, the incidence of GMH-IVH, particularly in its low-grade forms (Grade I and II), remains high, affecting around 44.68 percent of premature infants (Brouwer et al., 2014; Parodi et al., 2020). GMH-IVH continues to be a prevalent and clinically significant issue in preterm infants, especially those born extremely premature. The brains of extremely premature infants exhibit a higher germinal matrix volume, dense capillary networks, reduced support structures (Kidokoro et al., 2014), fewer anastomotic branches in the subependymal artery (Anstrom et al., 2002), oligodendrocytes are immature. Consequently, lower gestational ages render infants more vulnerable to the effects of cerebral blood perfusion and pressure, predisposing them to GMH-IVH (Gilard et al., 2020).GMH-IVH screening relies primarily on head ultrasound. Infants with low-grade GMH-IVH (grade I or II) are clearly at a much lower risk for developmental disabilities than infants with grade III GMH-IVH (Hambleton and Wigglesworth, 1976). Detecting microstructural damage to periventricular and subcortical white matter following low-grade GMH-IVH poses challenges; as such damage may be elusive. Furthermore, the reliability of head ultrasound in grading mild hemorrhages is limited, complicating the prediction of adverse outcomes (Leijser and de Vries, 2019). MR imaging is increasingly used in the extremely preterm population to complement cranial sonography to improve prognostic information and inform current clinical and future supportive care. It has been shown that low-grade GMH-IVH is followed by microstructural impairment, which are undetectable by ultrasound and conventional MRI. Consequently, there is a pressing need to identify new biomarkers for cerebral development in extremely preterm infants, offering potential for early detection and prevention of low-grade GMH-IVH.

Synthetic MRI, utilizing a multi-delay multi-echo sequence, emerges as a promising tool for simultaneously quantifying longitudinal relaxation time (T₁), transverse relaxation time (T₂), and proton density (PD). These parameters reflect microstructural changes during brain maturation, facilitating the objective assessment of abnormal maturation or disease conditions. Prior studies highlight the potential of T₁ and T₂ obtained from MRI at term-equivalent age (TEA) as imaging biomarkers for predicting neurodevelopmental outcomes in premature infants (Vanderhasselt et al., 2021; Dong et al., 2023). Synthetic MRI’s quantitative parameters effectively capture brain tissue characteristics and volume changes in premature infants with intraventricular hemorrhage, with T₁, T₂ relaxation times, and PD play crucial roles in diagnosing and evaluating intraventricular hemorrhage (Zhang et al., 2021). Synthetic MRI has shown promise in identifying early prognostic biomarkers for neurodevelopmental impairment, indicating its potential in predicting adverse outcomes linked to GMH-IVH (Vanderhasselt et al., 2021; Zhou et al., 2024). Despite these insights, there is not specifically explored the predictive performance of synthetic MRI metrics in assessing neurodevelopmental outcomes related to low-grade GMH-IVH in premature infants. This research gap underscores the necessity for further investigation into the potential of synthetic MRI metrics as predictive indicators in this specific population.

This study aimed to assess the diagnostic efficacy of synthetic MRI conducted at TEA around 38 weeks in forecasting adverse neurodevelopmental outcomes at 18 months among extremely preterm infants with low-grade GMH-IVH. Additionally, it sought to explore whether integrating relaxation times from various brain regions enhances predictive accuracy for neurodevelopmental outcomes.

This study delved into the diagnostic potential of synthetic relaxometry at TEA in predicting adverse neurodevelopmental outcomes among extremely preterm neonates with low-grade GMH-IVH at 18 months of corrected age. The combined T1-FWM + T1-Caudate + T2-FWM + T2-Caudate model demonstrated enhanced predictive power for adverse outcomes, surpassing the efficacy of individual data.

Our findings revealed a notable extension of T1, T2, and/or PD in the FWM and caudate in extremely preterm infants with neurodevelopmental disabilities compared to those without. The prolonged T1 and T2 relaxation times indicated changes in volumetric water content, compartmentalization, and macromolecular microstructures associated with early brain maturation and myelination (Vanderhasselt et al., 2021). This observation aligns with a recent study confirming similar results (Zhang et al., 2021). Hemorrhaging in the FWM leads to myelin breakdown and depletion, altering the myelin structure and resulting in elevated T1 and T2 values (Jara et al., 2022). Caudate involvement in germinal matrix hemorrhage often manifests as prolonged T1, T2, and PD, indicating delayed or impaired myelin development (Schmidbauer et al., 2021). -IVH characteristically initiates in the periventricular germinal matrix. The germinal matrix, located on the head of caudate nucleus and underneath ventricular ependyma, is a highly vascular collection of glial and neuronal precursor cells.

To substantiate the prognosis of neurodevelopmental outcomes, we delved into the MDI and PDI scores of extremely preterm neonates with low-grade GMH-IVH. Meicen et al. similarly identified a heightened risk of cognitive delay in these infants using the Bayley II assessment (Zhou et al., 2024), emphasizing the need for vigilant follow-ups in cases of mild IVH to detect potential developmental delays. However, no intergroup differences in PDI were found in this study. We speculate that the impairment of FWM and caudate in children with cerebral hemorrhage did not affect physical development.

Furthermore, our investigation unveiled elevated T1 and T2 relaxation times in the FWM and caudate among extremely preterm neonates with poorer cognitive scores. A noteworthy negative correlation emerged between T1 and T2 relaxation times in both the FWM and caudate and the MDI score. This aligns with a recent study, where higher T1 and T 2 relaxation times were observed in those with severe postnatal morbidities (Hagiwara et al., 2019). The caudate nucleus is intricately involved in working memory, cognitive function, and emotions. The association between higher T1 relaxation times in the caudate and lower MDI scores suggests a link between delayed myelin development in this region and impaired cognitive function in these neonates (Uccella et al., 2023). Similarly, the frontal lobe, responsible for high-level cognitive functions such as self-control, memory, and emotional expression (Passingham and Lau, 2023), displayed a negative correlation between T1 and T2 relaxation times in the FWM and MDI scores. This correlation implies that delayed myelin development in the FWM during the early neonatal period may negatively impact cognitive functions and vision at 18 months (Zhang et al., 2021; Zhao et al., 2021; Li et al., 2022).

In our study, we found that prolonged T1 and T2 relaxation times in the FWM and caudate serve as robust indicators of adverse neurodevelopmental outcomes. This highlights the vulnerability of these brain regions to cerebral hemorrhage. This affects the development of oligodendrocytes (Riddle et al., 2006). We hypothesize that the extension of relaxation time may signify disrupted cerebral maturation and reactive gliosis associated with low-grade GMH-IVH.

However, our study has several limitations. The relatively small sample size, although justified by the low incidence rate of low-grade GMH-IVH, necessitates cautious interpretation of the findings. Additionally, the neurodevelopmental follow-up was relatively short, confined to a subset of neonates at 18 months of age during the study period. Another constraint is the reliance on a single system for data collection. Future investigations should aim to gather data from diverse systems and time points to enhance both the volume and variety of information.

Our study illustrates the potential of predicting adverse neurodevelopmental outcomes in extremely preterm infants with low-grade GMH-IVH through synthetic MRI. The combination of the T₁-FWM + T₁-Caudate + T₂-FWM + T₂-Caudate model significantly improves the prediction of adverse outcomes compared to using individual data. These findings hold promise for assisting clinicians in establishing early intervention strategies in routine clinical practice.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving humans were approved by Ethics Committee of the Third Affiliated Hospital of Zhengzhou University. The studies were conducted in accordance with the local legislation and institutional requirements. Written informed consent for participation in this study was provided by the participants’ legal guardians/next of kin.

CZ: Writing – original draft. ZZ: Data curation, Formal Analysis, Writing – original draft. KW: Writing – review & editing. LW: Data curation, Formal Analysis, Writing – original draft. JL: Data curation, Formal Analysis, Writing – original draft. LL: Writing – review & editing. QX: Writing – review & editing. XW: Data curation, Formal Analysis, Writing – original draft. XiaZ: Funding acquisition, Supervision, Writing – review & editing. XinZ: Project administration, Validation, Visualization, Writing – review & editing.