Detecting neurodevelopmental trajectories in congenital heart diseases with a machine-learning approach

Elisa Cainelli1, Patrizia S Bisiacchi2,3, Paola Cogo4

  • 1Department of General Psychology, University of Padova, Padua, Italy. elisa.cainelli@unipd.it.

Scientific Reports
|January 29, 2021
PubMed

Insights

Children with congenital heart disease (CHD) show significant differences in cognitive and psychopathological domains compared to controls. Many patients exhibit ADHD-like profiles, highlighting the need for early and ongoing clinical attention.

Area of Science:

  • Neuroscience
  • Pediatric Cardiology
  • Developmental Psychology

Background:

  • Congenital heart disease (CHD) affects brain development.
  • Neuropsychological and psychopathological outcomes in CHD require further investigation.
  • Early cardiac surgery in CHD patients necessitates understanding long-term developmental impacts.

Purpose of the Study:

  • To characterize the neuropsychological and psychopathological profiles of children with CHD.
  • To identify potential clinical subtypes within the CHD population.
  • To explore associations between clinical parameters and neurodevelopmental outcomes in CHD.

Main Methods:

  • Prospective observational study of 74 children (37 with CHD, 37 controls) undergoing cardiac surgery before age five.
  • Comprehensive neuropsychological and psychopathological assessments conducted at least 18 months post-surgery.
  • Machine learning used for clustering and variable classification of psychopathological profiles.

Main Results:

  • Children with CHD demonstrated significant differences in intelligence, language, executive functions, and memory compared to controls.
  • Two psychopathological profiles emerged: a small group with high psychopathology symptoms and a larger group with ADHD-like features.
  • No significant associations were found between neuropsychological/psychopathological outcomes and the clinical parameters studied.

Conclusions:

  • Children with CHD exhibit substantial interindividual variability in neuropsychological and psychological outcomes.
  • ADHD-like profiles are common in CHD patients, often under-recognized by clinicians.
  • A lifespan approach is crucial for optimizing developmental trajectories in CHD patients due to ongoing brain maturation.