Alterations in white matter microstructure are associated with goal-directed upper-limb movement segmentation in children born extremely preterm

Abstract

Altered white matter microstructure is commonly found in children born preterm (PT), especially those born at an extremely low gestational age (GA). These children also commonly show disturbed motor function. This study explores the relation between white matter alterations and upper-limb movement segmentation in 41 children born PT (19 girls), and 41 children born at term (18 girls) at 8 years. The PT group was subdivided into extremely PT (E-PT; GA = 25-27 weeks, N = 10), very PT (V-PT; GA = 28-32 weeks, N = 13), and moderately PT (M-PT; GA = 33-35 weeks, N = 18). Arm/hand preference (preferred/non-preferred) was determined through object interactions and the brain hemispheres were designated accordingly. White matter alterations were assessed using diffusion tensor imaging in nine areas, and movement segmentation of the body-parts head, shoulder, elbow, and wrist were registered during a unimanual goal-directed task. Increased movement segmentation was demonstrated consistently on the preferred side in the E-PT group compared with the term born group. Also compared with the term born peers, the E-PT group demonstrated reduced fractional anisotropy (FA) in the cerebral peduncle (targeting the corticospinal tract) in the hemisphere on the non-preferred side and in the splenium of corpus callosum. In contrast, in the anterior internal capsule on the preferred side, the E-PT group had increased FA. Lower FA in the cerebral peduncle, but higher FA in the anterior internal capsule, was associated with increased movement segmentation across body-parts in a contralateral manner. The results suggest that impaired development of sensorimotor tracts in E-PT children could explain a sub-optimal spatiotemporal organization of upper-limb movements. Hum Brain Mapp 38:5051-5068, 2017. © 2017 Wiley Periodicals, Inc.

Keywords: anisotropy; corticospinal; diffusion tensor imaging; internal capsule; laterality; movement segmentation; sensorimotor.

Figure 1

Regions‐of‐interest (ROI) in the brain. The ROIs on the images have been outlined in a drawing software to facilitate viewing of the images, and they correspond to the actual ROIs outlined in the analyzing software. ROI sizes were constant for all brains, and the outlining was adapted to the particular brain in order to target the area in question. A: Middle cerebellar peduncle, 79 mm²; B: cerebral peduncle, 32 mm²; C: inferior longitudinal fascicle, 48 mm²; D: (i) genu of corpus callosum, 56 mm², (ii) anterior limb of the internal capsule, 38 mm², (iii) posterior limb of the internal capsule, 38 mm²; E: optic radiation, 48 mm²; F: (i) frontal white matter anterior, 56 mm², (ii) frontal white matter lateral 56 mm², (iii) splenium of corpus callosum, 56 mm²; G: truncus of corpus callosum, 96 mm²; H: superior longitudinal fascicle, 176 mm².

Figure 2

Scatter plot of the number of movement units versus fractional anisotropy. Data are from all participants and apply to the composite measure (shoulder, elbow, and wrist) when moving the preferred side (y‐axis) and the cerebral peduncle (corticospinal tract) on the non‐preferred side (x‐axis). MU, movement units; FA, fractional anisotropy; CST, corticospinal tract.