073 - White Matter Myelin Content in Preterm Infants Based on a Novel MRI Metric

Rose Horan, Stanford University School of Medicine, Bellingham, WA, United States; Rocio Velasco Poblaciones, Stanford University School of Medicine, Mountain View, CA, United States; Molly Lazarus, Stanford University School of Medicine, New York, NY, United States; Tiffany B. Sanchez Ramirez, Stanford University School of Medicine, East Palo Alto, CA, United States; Virginia A. Marchman, Stanford University School of Medicine, Stanford, CA, United States; Katherine Travis, Weill Cornell Medicine, White Plains, NY, United States; Heidi M. Feldman, Stanford University School of Medicine, Palo Alto, CA, United States

Background: Infants born very preterm ( < 32 wks gestational age (GA)) are at risk for white matter (WM) injury and dysmaturity, especially those with medical complications. The putative mechanism is oxidative-stress-induced injury to premature variants of myelin-producing cells, called pre-oligodendrocytes (pre-OLs), leading to hypomyelination. Pre-OLs are distributed non-uniformly throughout the brain; damage to myelin may be non-uniform. R1 is a novel MRI metric extracted from quantitative T1 (qT1) relaxometry, sensitive to myelin content in WM. We compared patterns of myelin content at near-term in infants born very preterm (PT) in relation to patterns previously found in infants born full term (FT) using R1 values.

Objective: To assess whether patterns of R1 in selected WM tracts on near-term MRIs are comparable among published values for FT infants and PT infants with low and high medical complexity.

Design/Methods: In this retrospective study, participants born < 32 weeks GA (n = 65, 28F, 37M) were scanned at postmenstrual age 34-42 wks using diffusion and qT1 MRI. The cohort was divided into 2 groups: low (LMC, N=36) and high medical complexity (HMC, N=29). Inclusion criteria for HMC were: inflammation (BPD, NEC, Sepsis), seizures, IVH>level 1, or ROP>level 2. Three tracts were selected for study based on a clear pattern of R1 in a study of FT infants: Corticospinal Tract (CST), Anterior Thalamic Radiation (ATR) and the Inferior Longitudinal Fasciculus (ILF). A one-sample t-test compared R1 values of all PT to reported FT values. A 2x3 ANCOVA with post-hoc tests analyzed main effects and interactions of LMC vs HMC and WM tract on R1 values, controlling for GA.

Results: Table 1 shows group characteristics; GA and birthweight were significantly lower in the HMC than LMC group. GA is controlled for in subsequent analyses. Figure 1 shows mean R1 values by group and tract. R1 values were lower in PT than FT. A significant main effect of tract revealed R1 values were different across tracts in LMC and HMC, following a similar pattern as in FT. Group differences between LMC and HMC were statistically significant in the ATR, and approaching significance in the ILF, controlling for GA. The group x tract interaction was not significant.

Conclusion(s): R1 values in the CST, ATR and ILF had a pattern preserved in the FT and both PT groups, indicating that the pattern of myelin maturation of WM circuits is retained in PT infants. The mean R1 tract values were higher in the LMC PT than the HMC PT in the ATR, suggesting higher myelin content in the low risk group in selected tracts.

Table 1. R1 values by white matter tract and level of medical complexity


Figure 1. R1 values by white matter tract and level of medical complexity