342 - Intermittent Hypoxemia (IH) in Preterm Infants: Glial Fibrillary Acidic Protein (GFAP) as a Potential Neuroinflammation-Related Brain Injury Biomarker

Publication Number: 342.5465

Elie G. Abu Jawdeh, University of Texas Southwestern, Dallas, TX, United States; Linda J. Van Eldik, University of Kentucky, Lexington, KY, United States; Richard Martin, Case Western Reserve University School of Medicine, Cleveland, OH, United States; Lina Chalak, UTSW, Dallas, TX, United States; Henrietta Bada, University of Kentucky College of Medicine, Lexington, KY, United States

Background: Intermittent Hypoxemia (IH) is a serious problem in preterm infants with growing evidence linking IH to brain injury and neonatal impairments. Additionally, previous studies suggest that IH may have proinflammatory effects (Darnall 2017, Abu Jawdeh 2020). While our recent findings indicate an association between IH and elevated S100B biomarker levels (PAS 2024, in prep.), early biomarkers for IH-related brain injury are still lacking, making it challenging to effectively monitor disease, optimize care, and predict outcomes.

Objective: To assess for the first time the relationship between IH and Glial Fibrillary Acidic Protein (GFAP), an injury biomarker particularly significant in neuroinflammation.

Design/Methods: We enrolled preterm infants ≤32 weeks gestational age (GA). To quantify an IH profile, oxygen saturation (SpO2) was collected using high-resolution (1Hz sampling, 2s averaging) pulse oximeters. The IH profile included measures such as frequency, severity, duration, and time in hypoxemia at multiple thresholds (SpO₂ < 85%, < 80%). To assess GFAP levels, urine samples were obtained at multiple time points during NICU stay, and GFAP measured by ultrasensitive Meso Scale Discovery assay. We excluded infants with severe intraventricular hemorrhage (contributor to brain injury), sepsis/necrotizing enterocolitis (contributors to inflammation/injury), or with no data at the time of GFAP. GFAP means were weighted based on the number of samples contributed by each patient. To better illustrate the relationship between IH and GFAP, we also stratified IH into tertiles based on IH frequency (low, moderate, high).

Results: A total of 29 infants were prospectively enrolled. After applying the exclusion criteria mentioned above, we analyzed 53 samples, with median GA of 28.3 weeks and a birth weight of 1060 g. A statistically significant correlation was observed between GFAP levels and IH profile (Table 1, all p< 0.05). Additionally, GFAP levels increased with higher IH tertiles (Figure; p = 0.009); i.e. the greater the IH burden the higher the GFAP level.

Conclusion(s): This study demonstrates an increase in GFAP, a potential neuroinflammation-related brain injury biomarker, with higher burdens of intermittent hypoxemia in preterm infants. This novel finding represents an important step toward developing early biomarkers for IH-related brain injury, which could help define injury thresholds, optimize care, and predict outcomes. A larger study is underway (K23HD109471) to further investigate these relationships.

Table 1: Correlations between GFAP and IH profile in Preterm Infants


Figure: Increased Glial Fibrillary Acidic Protein (GFAP) with higher IH burden
Increased GFAP with higher IH burden: Samples were divided into 3 tertiles based on the degree of IH burden: low, moderate (mod), and high IH frequency. There was an increase in GFAP (Red bars) with increasing IH frequency (Blue bars). There was a statistically significant difference across GFAP groups, p = 0.009 (Kruskal-Wallis). *p=0.026, **p=0.004 (Mann-Whitney). Tertile 1: Lowest IH frequency, Tertile 2: Moderate IH frequency, and Tertile 3: Highest IH frequency. There were also significant differences across IH groups, blue bars, all p<0.001 (asterisk not shown to avoid crowding). Red bars: GFAP levels. Blue bars: IH frequency. Data shown as Mean ± SEM.