Tractography-based quantitation of corticospinal tract development in premature newborns

Abstract

Objective: To evaluate the impact of early brain injury and neonatal illness on corticospinal tract (CST) development in premature newborns serially studied with diffusion tensor tractography.

Study design: Fifty-five premature newborns (median 27.6 weeks postmenstrual age) were scanned with magnetic resonance imaging (MRI) early in life and at term-equivalent age. Moderate-severe brain abnormalities (abnormal-MRI) were characterized by moderate-severe white matter injury or ventriculomegaly. Diffusion tensor tractography was used to measure CST diffusion parameters which reflect microstructural development: fractional anisotropy (FA) and average diffusivity (D(av)). The effect of abnormal-MRI and neonatal illness on FA and D(av) were assessed with multivariate regression for repeated measures adjusting for age at scan.

Results: Twenty-one newborns (38%) had abnormal-MRI on either scan. FA increased with age significantly slower in newborns with abnormal-MRI (0.008/week) relative to newborns without these MRI abnormalities (0.011/wk) (interaction term P = .05). D(av) was higher in newborns with abnormal-MRI (1.5 x 10(-5) mm(2)/sec; P < .001) for any given age at scan. In the 23 newborns (42%) with postnatal infection, FA increased more slowly (interaction term P = .04), even when adjusting for the presence of abnormal-MRI.

Conclusions: CST microstructural development is significantly impaired in premature newborns with abnormal-MRI or postnatal infection, with a pattern of diffusion changes suggesting impaired glial cell development.

Figure 1. Diffusion tensor tractography of the corticospinal tract in a premature newborn studied serially

Representative images from a premature newborn delivered at 27 weeks gestational age and studied with MRI at: (A) 29 weeks of postmenstrual age (B) and 45 weeks postmenstrual age. (Top) Axial diffusion tensor imaging encoded anisotropy color map illustrating the region of interest in the posterior limb of the internal capsule (in white). The color convention used to display the predominant diffusion direction has red representing right–left, green representing anterior-posterior, and blue representing superior-inferior anatomical directions (2). (Bottom) 3D renderings of the DTT delineated corticospinal tract (CST) in red. A seeding ROI was first placed at the posterior limb of the internal capsule (Top, in white) and then limiting ROIs were placed at the precentral gyrus and cerebral peduncle to define the tract. Diffusion parameters were averaged over the entire length of the tract.

Figure 2. Developmental trajectory of CST diffusion parameters obtained from serial DTT scans in premature newborns with and without moderate to severe MRI abnormalities (abnormal-MRI) and in those with or without postnatal infections

A. The rate of change of fractional anisotropy is significantly different in infants with and without abnormal-MRI (significant interaction). Fractional Anisotropy increases less rapidly in newborns with abnormal-MRI (interaction term P=0.05). B. Average Diffusivity is higher in newborns with abnormal-MRI (P=0.2). The rate of change of average diffusivity is not significantly different. C. Axial diffusivity is not significantly different in newborns with abnormal MRI (P=0.7). D. Radial diffusivity is higher in newborns with abnormal MRI (P=0.1). E. Fractional Anisotropy increases less rapidly in newborns with infection (interaction term P=0.04). F. Average Diffusivity decreases less rapidly in newborns with infection (interaction term P=0.07).