Functional Connectivity Relationships to Longitudinal Motor Outcomes Differ in Very Preterm Children With and Without Brain Injury.

Background and objectives: Children born very preterm (VPT) have high rates of motor disability, but mechanisms for early identification remain limited, especially for children who fall behind in early childhood. This study examines the relationship between functional connectivity (FC) measured at term-equivalent age and motor outcomes at 2 and 5 years.

Methods: In this longitudinal observational cohort study, VPT children (gestational age 30 weeks and younger) with and without high-grade brain injury underwent FC MRI at term-equivalent age. Motor development was assessed using the Bayley Scales of Infant Development, Third Edition, at corrected age 2 years and Movement Assessment Battery for Children, Second Edition, at age 5 years. Logistic and negative binomial/Poisson regression models examined relationships between FC measures and 5-year task scores, with and without 2-year scores as covariates. Infants were categorized as "injured" or "uninjured" based on structural MRI findings at term-equivalent age.

Results: In the injured group (n = 34), each 1 SD decrease in neonatal left-right motor cortex FC was related to approximately 4× increased odds of being unable to complete a fine motor task at age 5 (log odds = -1.34, p < 0.05). In the uninjured group (n = 41), stronger basal ganglia-motor cortex FC was related to poorer fine motor scores (Est = -0.40, p < 0.05) and stronger cerebellum-motor cortex FC was related to poorer balance and fine motor scores (Est = -0.05 to -0.23, p < 0.05), with balance persisting with adjustment for 2-year scores.

Discussion: In VPT children with brain injury, interhemispheric motor cortex FC was related to motor deficits at 5-year assessment, similar to previous findings at 2 years. In uninjured children, FC-measured disruption of the motor system during the neonatal period was associated with motor planning/coordination difficulties that were not apparent on 2-year assessment but emerged at 5 years, suggesting that the neural basis of these deficits was established very early in life. Subsequently, 2-year follow-up may not be sufficient to detect milder motor deficits in VPT children, and they should be monitored for motor difficulties throughout the preschool years. For all VPT children, FC at term-equivalent age has the potential to improve our ability to predict disability before it presents behaviorally.