Brain Network Characterization of Preterm Infants With Bronchopulmonary Dysplasia.

ABSTRACT

BACKGROUND: Bronchopulmonary dysplasia (BPD) affects the microstructure of white matter in preterm infants, but its influence on the changes of the brain structural network has not been elaborated. This study aims to investigate the connectivity characteristics of the brain structural network of BPD by using diffusion tensor imaging. METHODS: Thirty-three infants with BPD and 26 infants without BPD were enrolled in this study. Brain structural networks were constructed utilizing automated anatomic labeling mapping by tracing the fibers between each pair of regions in individual space. We calculated network metrics such as global efficiency, local efficiency, clustering coefficients, characteristic path length, and small-worldness. Then we compared the network metrics of these infants with those of 57 healthy term infants of comparable postmenstrual age at magnetic resonance imaging scan. Finally, network-based statistics was used to analyze the differences in brain network connectivity between the groups with and without BPD. RESULTS: Preterm infants with BPD had higher local efficiency and clustering coefficient, lower global efficiency, and longer characteristic path length. Also, preterm infants with BPD had decreased strength of limbic connections mainly in four brain regions the left lingual gyrus, the left calcarine fissure and surrounding cortex, the right parahippocampal gyrus, and the left precuneus. CONCLUSIONS: Our findings suggest that preterm infants with BPD have lower network integration and higher segregation at term-equivalent age, which may reflect a compensatory mechanism. In addition, BPD affects brain regions involved in visual as well as cognitive functions; these findings provide a new approach to diagnose potential brain damage in preterm infants with BPD.