Deviation from normative volumetric development in the preterm brain is associated with childhood psychopathology

• In Event: Poster Session 03
  In Poster Session: PS 03 Topic - Biological Processes: Neuroscience and Genetics

Abstract

Preterm birth is associated with a range of adverse mental health outcomes throughout development. However, there is considerable heterogeneity in the extent and type of mental health problems experienced by individuals born preterm, and the exact mechanisms underlying specific outcomes remain unclear. Here, we assess individual deviations from normative trajectories of regionally specific volumetric brain development in the neonatal preterm brain, and relate these deviations to dimensional measures of psychopathology in childhood.
Eighty-nine preterm infants (born < 33 weeks of gestation) underwent magnetic resonance imaging (MRI) at term-equivalent age as part of the Evaluation of Preterm Imaging (EPrime) study. Typical volumetric brain development was characterised using Gaussian process regression on MRI scans of 275 healthy term-born infants collected as part of the developmental Human Connectome Project (dHCP), modelling the effects of age at scan and sex. For each preterm infant, we then calculated the deviation from normative growth of the volumes of white matter, cortical grey matter, and striatal structures. Deviation z-scores were defined as the difference between the observed and model-predicted regional volume given each infant’s age and sex, normalised by the prediction uncertainty. At a median age of 4.9 years (range: 4.2-7.1 years), psychopathology was assessed via parent-report measures of temperament, emotional and behavioural problems, and psychiatric symptoms. The latent structure of these measures was captured using a bifactor model, which resulted in one general psychopathology (‘p’) factor and three independent specific factors reflecting inattention/hyperactivity, socioemotional problems, and autism spectrum symptoms. We tested for significant associations between each of these factors and regional deviation from normal volumetric development, controlling for gestational age.
Reduced volume of the left caudate relative to typical development at term equivalent age was significantly associated with increased general psychopathology in childhood (β = -0.15, p = .007). In contrast, reduced global white matter volume at term was associated with an increase on the inattention/hyperactivity factor in childhood (β = -0.07, p = .008). These results provide evidence for region and tissue specific links between brain development and separable facets of childhood psychopathology, potentially providing an explanatory framework for the wide range of outcomes in preterm cohorts. Crucially, this method allows for subject-level characterisation of regional brain development relative to normative development, rather than relying on absolute volumetric differences, providing a more nuanced approach to predicting psychopathology associated with preterm birth.

Authors

• Lucy Denise Vanes, King's College London
  Presenting Author

• Ralica Dimitrova, King's College London
  Non-Presenting Author

• Jonathan O'Muircheartaigh, King's College London
  Non-Presenting Author

• Shona Falconer, King's College London
  Non-Presenting Author

• Dafnis Batalle, King's College London
  Non-Presenting Author

• Jo Hajnal, King's College London
  Non-Presenting Author

• Serena Counsell, King's College London
  Non-Presenting Author

• David Edwards, King's College London
  Non-Presenting Author

• Chiara Nosarti, King's College London
  Non-Presenting Author