Functional Connectivity in Adolescents with Congenital Heart Disease or Neonatal Encephalopathy

1261 

Abstract Submission 

Megan Martin¹, Lauren Harasymiw¹, Shabnam Peyvandi¹, Elizabeth George¹, Patrick McQuillen¹, Duan Xu¹

¹University of California, San Francisco, San Francisco, CA

Megan Martin  
University of California, San Francisco
San Francisco, CA

Lauren Harasymiw  
University of California, San Francisco
San Francisco, CA

Shabnam Peyvandi  
University of California, San Francisco
San Francisco, CA

Elizabeth George  
University of California, San Francisco
San Francisco, CA

Patrick McQuillen  
University of California, San Francisco
San Francisco, CA

Duan Xu  
University of California, San Francisco
San Francisco, CA

Although congenital heart disease (CHD) and hypoxic ischemic encephalopathy (HIE) are different pathological conditions, CHD restricts oxygen delivery to the neonatal brain and HIE results from restricted oxygen delivery to the neonatal brain. Both are associated with high incidence of neonatal brain injury and altered brain development. Prior studies have shown differences between these groups in brain size and microstructural development as well as associations between graph theory metrics and intellectual performance derived from functionally connected neural networks using resting state functional MR images (rsfMRI). A negative association between average pathlength and intellectual performance and a positive relationship between global efficiency and intellectual performance have been observed. In this study, we compare graph theory metrics of functionally connected neural networks of adolescents who were born with CHD or HIE to examine similarities and differences.

We evaluated rsfMRI in adolescents with CHD (n=23) between the ages of 8 and 19 years old and HIE (n=20) between the ages of 10 and 15 years old. These adolescents were enrolled in respective longitudinal cohort studies as neonates. An fMRI sequence with 3mm resolution, a repetition time of 2 seconds and 200 passes was acquired continuously on a 3T GE MR750 scanner (GE Healthcare, Waukesha, WI, USA) while the subjects were at rest. Additionally, subjects underwent detailed neuropsychological testing. The CONN toolbox (version 22a) was used to preprocess and denoise the rsfMRI data, co-register with the T1 weighted anatomical image and MNI adult brain atlas, and compute functional brain connectomes, adjacency matrices, and graph theory metrics. The graph theory metrics included in this analysis were global and local efficiency, betweenness, closeness centrality, eigenvector centrality, eccentricity, cost, average pathlength, clustering coefficient, and degree. An ANCOVA test was used to assess age as a covariant of neural network graph theory metrics and evaluate groupwise significant differences between CHD and HIE. Groupwise mean and standard deviation were computed when a significant difference was present in a network graph theory metric.

With age evaluated as a covariant, significant differences by diagnosis were identified in two or more graph theory metrics of the sensory motor, visual, salience, dorsal attention, language, and cerebellar networks. No significant differences by diagnosis were identified in the default mode or frontoparietal networks. When significantly different, mean average pathlength was lower in the CHD cohort than in the HIE cohort, except in the cerebellar posterior network, where it was lower in the HIE cohort (Figure 1). Conversely, when significantly different, global efficiency was generally higher in the CHD cohort than in the HIE cohort, except in the dorsal attention frontal eye field and the cerebellar posterior networks where this metric was higher in the HIE cohort (Figure 1).

Our study suggests differences in functional connectivity of neural networks between adolescents with CHD as compared to those with HIE. Further work is needed to determine the association of these differences with detailed cognitive or functional outcomes available for these cohorts.

Early life, Adolescence, Aging ¹

fMRI Connectivity and Network Modeling ²

Congenital

Development

DISORDERS

FUNCTIONAL MRI

MRI

Neurological

PEDIATRIC