Metabolic maturation of white matter is altered in preterm infants

Abstract

Significant physiological switches occur at birth such as the transition from fetal parallel blood flow to a two-circuit serial system with increased arterial oxygenation of blood delivered to all organs including the brain. In addition, the extra-uterine environment exposes premature infants to a host of stimuli. These events could conceivably alter the trajectory of brain development in premature infants. We used in vivo magnetic resonance spectroscopy to measure absolute brain metabolite concentrations in term and premature-born infants without evidence of brain injury at equivalent post-conceptional age. Prematurity altered the developmental time courses of N-acetyl-aspartate, a marker for axonal and neuronal development, creatine, an energy metabolite, and choline, a membrane metabolite, in parietal white matter. Specifically, at term-equivalency, metabolic maturation in preterm infants preceded development in term infants, but then progressed at a slower pace and trajectories merged at ≈340-370 post-conceptional days. In parieto/occipital grey matter similar trends were noticed but statistical significance was not reached. The timing of white matter development and synchronization of white matter and grey matter maturation in premature-born infants is disturbed. This may contribute to the greater risk of long-term neurological problems of premature infants and to their higher risk for white matter injury.

Figure 1. MR images and ¹H MR spectra of term-born and premature-born infants.

T2-weighted MR imaging of the newborn brain indicating the regions of interest in parietal WM, parieto/occipital GM, and frontal WM from where spectra were acquired for this study (A). In representative parietal WM spectra of term-born and premature-born infants at equivalent post-conceptional (PC), NAA and Cr are visibly more prominent in the preterm spectrum (B). When data from all infants are analyzed, Cho is also significantly different between the two groups.

Figure 2. NAA concentrations versus post-conceptional age.

Shown are NAA concentrations versus PC age of term newborns in parietal WM, parieto/occipital GM, and frontal WM with the curves that provided the best fits superimposed to the data (A). Note the different shape of the curves. Whereas the time of the fastest increase of NAA in parietal WM is around 295 PC days, the fasted increase of NAA in GM and frontal WM is observed 2–3 months later at around 350 and 390 PC days. NAA is believed to indicate axonal growth and neuronal maturations and these data are consistent with developmental processes in parietal WM preceding changes in GM or frontal WM. When NAA concentrations of premature infants are plotted (B), a discrepancy is readily noticed for parietal WM (note that the fitted curve for term infants was carried over from figure A and that there are more points above than below that curve) but is not observed for parieto/occipital GM and is also less apparent for frontal WM where fewer data points are available. When the normalized difference (z-score) is analyzed, the majority of the preterm data in parietal WM are positive, i.e. at equivalent PC age concentrations are higher in premature brain. However, the differences in parietal WM decrease with increasing PC age. Indeed, at around 340 PC days there is no difference between term and premature brain in respect to NAA concentrations in parietal WM. It appears that frontal WM follows parietal WM. Albeit the mean difference between term and preterm infants is not significant, there is a significant negative slope comparable with the observation made for parietal WM (Tab. 2). PC = post-conceptional.

Figure 3. Creatine concentrations versus post-conceptional age.

Unlike for NAA, where parietal WM shows a significant different time course when compared with GM and frontal WM, the times courses for the three brain regions are very comparable (A). Similar with the observations for NAA, Cr concentrations in premature infants are generally higher than in term-born infants, particularly in parietal WM (B). As for NAA, the difference between term and preterm infants decreases with age (C).

Figure 4. Concentrations of myo-inositol and glutamate versus post-conceptional age.

This figure shows the best fit (solid line) of the term data for mIns and glutamate (Glu) with the individual measurements in preterm infants superimposed in parietal WM. There are no apparent differences between term and preterm infants at equivalent PCA. These graphs should be compared with Figs. 2B and 3B (left columns) that illustrate the differences observed for NAA and Cr in parietal WM.