140 - Multimodal Neuroimaging Demonstrates Hyperconnectivity in Extremely Preterm Children at School Age: An Exploratory Study Investigating the Home Literacy Environment and Digital Media Use

Sunday, April 30, 2023

3:30 PM – 6:00 PM ET

Poster Number: 140
Publication Number: 140.362

Kaitlyn Kelly, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, United States; John S. Hutton, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, United States; Darren S. Kadis, The Hospital for Sick Children, Toronto, ON, Canada; Nehal A. Parikh, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, United States; Maria E. Barnes-Davis, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, United States

Presenting Author(s)

Kaitlyn Kelly, MD (she/her/hers)

Fellow
Cincinnati Children's Hospital Medical Center
Cincinnati, Ohio, United States

Background:

Extremely preterm (EPT, < 28 weeks) children are at risk for aberrant language development. On fMRI-constrained magnetoencephalography (MEG), EPT exhibit functional hyperconnectivity related to language gains (Barnes-Davis 2021). We have shown language skills are associated with home literacy environment and screen time in term children (TC) (Hutton 2020).

Objective:

Observational pilot study investigating the relationship of screen time and home literacy environment to functional brain connectivity and language outcomes in EPT and TC.

Design/Methods:

23 children (8 EPT, 15 TC) aged 4-6 years were prospectively enrolled and completed validated measures of general abilities (WNV); vocabulary (PPVT, EVT); shared reading (DialogPR); home literacy environment (StimQ2-P); and digital media use (ScreenQ). Group differences were assessed via t-tests (p< 0.05). A stories-listening task was performed during fMRI (Philips Achieva 3.0T scanner) and MEG (275-channel CTF system at 1200 Hz) per prior report (Barnes-Davis 2021). A joint fMRI activation map was generated and parcellated into 200 random units. Parcels with 25% activated voxels served as nodes for MEG analysis (Fig 1). Functional connectivity between nodes was evaluated with debiased weighted phase lag index. Frequency bands with group differences were identified and tested for significant subnetworks in Network Based Statistics (NBS).

Results:

On preliminary analysis of the first 23 children enrolled, significant group differences were found in race, income, and parental education (Table 1). EPT and TC scored within normal range without group differences on WNV or PPVT, but with significant differences on EVT (p< 0.05). No significant group differences were found in DialogPR, StimQ2-P, or ScreenQ. On fMRI, EPT and TC demonstrated task-specific bitemporal activation without significant differences. On MEG, EPT demonstrated regional hyperconnectivity compared to TC (9.5-17 Hz). NBS identified a significant subnetwork contributing to observed functional hyperconnectivity (5000 permutations, median t threshold 1.4, p< 0.05, Fig 2). Recruitment of the full sample (n=35, 15 EPT, 20 TC) and analyses relating brain connectivity to home literacy environment and digital media use are ongoing.

Conclusion(s):

We replicated MEG hyperconnectivity in EPT versus TC in a new sample of 4-6 year olds. Previously, hyperconnectivity has been correlated to language skills. Following completion of recruitment of a balanced sample, we will explore the relationship of the hyperconnectivity in EPT and measures of shared reading, home literacy environment, and digital media use.