Substructural Topographic Map of the Interhemispheric-Cortices Connectivity in Neonate Boys And Girls

Purposes: The corpus callosum (CC) is the largest bundle of white matter fibers interconnecting the 2 hemispheres. The development starts at 14 weeks of gestation and there is a posterior (splenium) to anterior (genu) gradient of myelination, although the genu grows prior to the splenium. Abnormal CC development has been observed in preterm infants [1], low birth weight children [2], and neonates with congenital heart diseases [3]. Poor neuro-development outcomes has been correlated with altered CC microstructures in periventricular leukomalecia [4] and preterm infants [5]. Using fiber tracking we aim to draw a topographic map of the interhemispheric fibers in healthy neonates by segmenting the corpus callosum in 5 sub-structures: genu, rostrum, body, isthmus and splenium. Material et Methods: DTI was performed on 3T scanner with 8-channel head coil and 35 non-collinear diffusion gradient directions plus one T2W volume and double refocusing pulses to reduce eddy-current artifacts. The diffusion sensitivity was set to b=700 (s/mm), pixel = 0.85x0.85 mm and slice thickness = 2.5 mm. The CC was manually segmented in 5 substructures on diffusion color-encoded map by measuring the anterior-to-posterior length on sagittal view. The genu, most anterior substructure, was set to 1/6 of the entire length, followed by the rostrum 2/6 of the length, then the body which equals 4/6, then the isthmus which measures 4/5 and finally the splenium, the most posterior substructure (Figure 1). Twenty healthy term infants were recruited form the postnatal ward at the University Hospital of Zurich (Table 1). Fiber tracking was carried out on DTIStudio with FACT algorithm and the following criteria FA = 0.15, and angulation < 60°. Boys versus girls statistical analysis (multi-comparison analysis of covariances) of fiber length (mm), anisotropy (FA), and mean (MD), radial (E23) and axial (E1) diffusions was performed with age at MRI as covariate.