Alexander Weber

Background: Preterm birth is associated with a higher rate of brain injury and neurodevelopmental delays in children. The brain is dependent on oxidative metabolism of glucose to meet its significant energy requirements. Understanding the cerebral metabolic rate of oxygen (CMRO₂) and how it is affected by ventilation in the neonatal population is crucial to the advancement of neonatal medicine. This study aims to assess the feasibility of quantifying CMRO₂ and cerebral blood flow (CBF) using magnetic resonance imaging (MRI) in preterm infants at term equivalent age (TEA) and to determine how various ventilatory support may affect CMRO₂. We hypothesized that increased time on mechanical ventilation would be negatively associated with CMRO2 and CBF levels in preterm neonates at TEA.Methods: Very preterm neonates (n=19) born 32 weeks gestational age were recruited in the NICU and scanned at TEA. The scanning sequence consisted of a 3D T1-weighted, 3D T2-weighted, arterial spin labeling (ASL), and susceptibility weighted imaging (SWI) scan. CBF maps were calculated with ASL, and SWI was used to create quantitative susceptibility maps (QSM) to calculate the cerebral venous oxygen saturation (CSvO₂). Clinical data was acquired from the Canadian Neonatal Network (CNN) database and chart review. Results: Average global CMRO₂ was 28.49±3.29 µmol/100g/min (mean±SD), consistent with values of CMRO₂ reported in the literature falling within the range of 25-45 µmol/100g/min. The average global CBF was 13.96±4.22 mL/100g/min. CMRO₂ and CBF were found to be positively correlated with days on noninvasive ventilation and negatively correlated with days on room air. CBF was also positively correlated with total time on ventilation.Conclusion: We demonstrated the feasibility of noninvasive quantification of CMRO2 and CBF using MRI in very preterm neonates at TEA. The levels of CMRO2 and CBF found in these infants were consistent with values previously reported for neonates of similar sample populations. Higher CMRO2 values with longer durations of noninvasive ventilation may indicate better health outcomes. Previous literature has shown no adverse effects of prolonged noninvasive ventilation, suggesting that invasive ventilation should be replaced by noninvasive ventilation when possible.

View record

Interrupted brain maturation from preterm birth increases the consequences of altered functional development ( 37 weeks gestational age (GA)). Mono-fractal analysis is an advanced functional magnetic resonance imaging (fMRI) analysis method that calculates the temporal auto-correlation characteristics from the blood oxygenation level dependent (BOLD) signal. This can be measured using the Hurst exponent (H). Our study goal was to measure H of the resting state-fMRI (rs-fMRI) BOLD signal to investigate alterations in brain signaling complexity at preterm age, term equivalent age and term healthy controls scanned at term age. Participant rs-fMRI and diffusion tensor imaging (DTI) data from 716 neonates born 23 to 43 weeks GA were obtained from the Developing Human Connectome Project. From the rsfMRI scans, independent component analysis was used to identify 13 resting state networks (RSNs). Temporal complexity was determined using H calculated from the power spectral density of the BOLD signal using Welch’s method. DTI scans from the preterm cohort were used to indirectly measure myelination with fractional anisotropy (FA) and radial diffusivity (RD) measurements. H in brain tissues and RSNs were assessed across gestational age and scan age, and Pearson's correlation was conducted between H and DTI measures. H significantly increased from preterm to TEA assessment, and earlier birth age contributed to lower H values. H begins below 0.5 at preterm age and crosses 0.5 at around term age in most regions. Motor and sensory networks were found to have the greatest increase in H. Correlations between DTI measures and H were moderate but significant, demonstrating somewhat parallel development of structural and functional systems. We found that H was an indicator for BOLD signal processing maturation in the infant brain. Accordingly, preterm infant signaling transforms from anticorrelated to correlated but is reduced by preterm birth compared to term born infants.

View record