36 - Associations among EEG power, gestational and postnatal age, and later neurodevelopment of preterm infants born before 29 weeks gestation

Monday, May 1, 2023

9:30 AM – 11:30 AM ET

Poster Number: 36
Publication Number: 36.435

Soujin Choi, Northwestern University, Evanston, IL, United States; Erin Lonergan, Ann & Robert H. Lurie Children's Hospital of Chicago, Palos Heights, IL, United States; Allaa Fadl-Alla, Ann & Robert H. Lurie Children's Hospital of Chicago, Chicago, IL, United States; Casey Rand, lurie childrens hospital and stanley manne research center, Chicago, IL, United States; Brad Hopkins, Ann & Robert H. Lurie Children's Hospital of Chicago, Douglas, MI, United States; Narayanan Krishnamurthi, Lurie Children's Hospital, Chicago, IL, United States; Lynn D. Boswell, Ann & Robert H. Lurie Children's Hospital of Chicago, Chicago, IL, United States; Aaron Hamvas, Lurie Children's Hospital / Northwestern University, Chicago, IL, United States; Raye-Ann deRegnier, Ann & Robert H. Lurie Children's Hospital of Chicago, Chicago, IL, United States; Debra E. Weese-Mayer, Ann & Robert H. Lurie Children's Hospital of Chicago, Chicago, IL, United States; Elizabeth S.. Norton, Northwestern University, Evanston, IL, United States

Presenting Author(s)

Elizabeth S. Norton, PhD (she/her/hers)

Associate Professor
Northwestern University
Evanston, Illinois, United States

Co-Author(s)

SC

Soujin Choi, MA (she/her/hers)

Northwestern University
Evanston, Illinois, United States

Background: Preterm birth is associated with atypical neural activity assessed via electroencephalography (EEG) concurrently and over time. Children born preterm are at substantially elevated risk for neurodevelopmental impairments, in cognition and notably in language. Few studies have assessed how EEG relates to early brain maturation and whether it could indicate a later risk for neurodevelopmental impairments.

Objective:
This study aimed to assess how various bands of EEG power among children born extremely preterm related with gestational age (GA) and postmenstrual age (PMA) at assessment, as well as neurodevelopment in the first year.

Design/Methods: For 12 infants in the Pre-Vent study (NIH U01HL133704) born at 24-28 weeks GA, EEG was recorded in the NICU via SomnoStar system at 6 scalp sites during a quiet awake state at PMA of 35-38 weeks. EEG was processed via a standard research processing pipeline (EEGLab/ERPLab). Relative power was calculated for frequency bands delta, theta, alpha, beta, and gamma at literature-defined scalp sites. Neurodevelopment (language, cognition) was clinically assessed via the Bayley Scales of Infant and Toddler Development at 6-12 months corrected age.

Results:
Spearman correlation coefficients and p-values are reported in Tables 1-2. GA was positively related to alpha and theta, and negatively related to beta and delta (medium-large effect sizes, rs >+/-.42). In relation to PMA at EEG collection, negative correlations were observed, large effects with delta (p < .05), and medium with alpha and theta; there was a medium positive correlation with beta. Gamma was not related to age.

We used partial Spearman correlations accounting for GA to assess how EEG predicted later language and cognitive standard scores. Language and cognition were correlated at r=.70 and showed generally similar patterns with EEG; both had a large positive correlation with delta (p < .05 for language); theta and beta had small to medium negative correlations with both. Gamma had a positive medium relation with cognition but not language.

Conclusion(s):
This preliminary study suggests that EEG power analysis may be useful for examining brain maturation and predicting neurodevelopment of preterm infants. The early differences in EEG associated with GA likely reflect brain maturation in utero which may differ from those that promote neurodevelopment. The associations between neurodevelopment and EEG power were similar to a previous term newborn study (Brito et al., 2016), suggesting the importance of considering the individual differences in neonatal EEG, as will be a focus in future analyses from this study.