Related Experiment Video
Updated: Sep 5, 2026
Assessment and Evaluation of the High Risk Neonate: The NICU Network Neurobehavioral Scale
Published on: August 25, 2014
Relationship Between Bronchopulmonary Dysplasia, Corpus Callosum Microstructure at Term-Equivalent Age, and
Kylan A Nelson1, Katsuaki Kojima2, Anoosha Sri1
1Department of Pediatrics, University of Cincinnati College of Medicine, 3230 Eden Avenue, Cincinnati, OH 45267.
Objective:
To describe changes in fractional anisotropy (FA) of the corpus callosum and neurodevelopmental outcomes in infants born preterm at or less than 32 weeks' gestational age (GA) with bronchopulmonary dysplasia (BPD).
Study Design:
In this prospective regional cohort study, 392 infants born preterm underwent diffusion MRI (dMRI) at term corrected age (CA) and were invited for standardized motor, cognitive, and language developmental testing (Bayley-III) at 2 years CA. We used an automated tractography software to segment the corpus callosum (CC) into its 7 subsegments and calculate fractional anisotropy (FA), a measure of dMRI brain microstructure. We performed multivariable linear regression analyses to examine associations between BPD severity and FA values, and between BPD and Bayley-III scores, adjusting for confounders. Mediation analysis assessed whether FA values mediated the relationship between BPD and Bayley scores.
Results:
Of 392 infants, 227 (57.9%) had no BPD, 91 (23.2%) grade 1, 56 (14.3%) grade 2 and 18 (4.6%) grade 3. Of the same 392 infants, 245 had usable dMRI scans and 341 (87%) returned for Bayley-III testing. In multivariable analyses, higher BPD grades were associated with lower FA in the posterior midbody and isthmus of the CC (both Q≤0.005). Infants with higher grades of BPD also had lower Bayley-III cognitive (P=0.016), language (P=0.019), and motor (P<0.001) scores. Corpus callosum microstructure partially mediated the association between BPD severity and cognitive outcome (posterior midbody, 12.7% of the total effect; P = 0.026), although mediated effects were modest and most did not reach statistical significance. The direct effect of BPD persisted throughout.
Conclusions:
In our regional preterm cohort, greater BPD severity was linked to poorer neurodevelopment, and this relationship was partially mediated by aberrant CC development, although the mediation was modest.
