370 - Association of Umbilical Cord Blood pH and Apgar Scores with Resting State Functional Brain Connectivity of Neonates with Hypoxic-ischemic Encephalopathy.
Friday, April 25, 2025
5:30pm – 7:45pm HST
Publication Number: 370.3914
Sabrina Mastroianni, McMaster University Michael G. DeGroote School of Medicine, Thornhill, ON, Canada; Anagha Vinod, Mcmaster University, Hamilton, ON, Canada; Naiqi G. Xiao, McMaster University, Hamilton, ON, Canada; Heather Johnson, McMaster University Michael G. DeGroote School of Medicine, Hamilton, ON, Canada; Ipsita Goswami, McMaster University Michael G. DeGroote School of Medicine, Hamilton, ON, Canada
Medical Student (MD, c2026) McMaster University, Michael G. DeGroote School of Medicine Thornhill, Ontario, Canada
Background: Hypoxic-ischemic injury to the neonatal brain evolves through a secondary phase of neuronal death followed by a tertiary phase of reorganization and recovery. The resting state functional connectivity of the cerebral cortex in neonates with hypoxic-ischemic encephalopathy (HIE) who undergo hypothermia therapy is largely unknown. Objective: To investigate the association between umbilical cord pH and Apgar scores at birth with resting-state functional connectivity (rsFC) patterns during the tertiary phase of HIE. Design/Methods: Neonates >35 weeks gestation and birthweight >1800g with fetal acidosis and encephalopathy were prospectively enrolled. A wearable multichannel functional near-infrared spectroscopy (fNIRS) suite with a neonatal cap that holds 20 emitters and 16 detectors was used, creating 27 channels per side covering frontal, parietal, and occipital lobes. Measurements were taken during sleep or in a quiet alert state for 20 minutes between 7-15 days of life in a dark space within the neonatal intensive care unit. Graph theory metrics (Assortativity, Hierarchy, global and local efficiency, and small worldness) were calculated. Assortativity is the tendency of nodes to connect to other nodes with similar degrees. Hierarchy refers to a layered structure with more connected nodes at the top and less connected nodes below. The shortest path length in a network is the minimum number of edges needed to travel from one node to another. Results: Of the 8 subjects, 5 were males, and 4 were born by C-section, with mean (SD) gestation and birthweight of 38(2) weeks and 2981(225)g, respectively. Stage of encephalopathy at admission was mild (n =2) and moderate (n=6). The median day of fNIRS measurement was 8 days of life. Arterial cord blood pH had a negative correlation with Hierarchy [r -0.85, p-value 0.03]. Apgar score at 10 mins had a negative correlation with Assortativity [r -0.93, p-value 0.006] and average shortest path length [ r -0.81, p-value 0.04].
Conclusion(s): More severe perinatal insult is associated with brain networks with (i) high Assortativity, which enhances localized information sharing and resilience even if parts of the network may be damaged, (ii) more hierarchical networks that facilitate adaptive processes, allowing for failure of lower-level nodes without disrupting the network’s overall functionality, and (iii) higher shortest path length indicating decreased efficiency. Such a network supports efficient local processing within modules and specialized functions but cannot quickly integrate information across the entire network.
Table 1: Correlation of network metrics and clinical variables in neonatal hypoxic ischemic encephalopathy Table 1.pdf
Figure 1: Network metrics of individual participants in the study Metrics Figures.pdf