Association between Biomarkers of Brain Damage, Inflammation, and Neurological Outcomes in Asphyxiated Infants Treated and Untreated with Therapeutic Hypothermia

Introduction: Asphyxia induces an acute inflammatory response in injured brain that contributes to tissue damage and neurologic dysfunction. When the brain is subjected to trauma, the intracerebral activation of the complement cascade occurs. It has been shown that in animal models, genetic or pharmacological inhibition of terminal complement components decreases brain neutrophil accumulation, blood–brain barrier damage, apoptosis, and neurologic dysfunction. Mannose-binding lectine (MBL) initiates complement activation and could exacerbate tissue damage after systemic ischemia/reperfusion insult. The S100B protein is a marker of tissue damage after traumatic brain injury. There is evidence supporting its prognostic ability as a long-term predictor of postasphyxia brain dysfunction. We tested the hypothesis that therapeutic hypothermia, currently the treatment of choice for acute asphyxia in at term infants, could influence the release of inflammatory mediators, including MBL, and the release of S100B protein from damaged brain cells in asphyxiated neonates. We aimed at measuring serum levels of MBL and S100B in asphyxiated neonates both during total body TH and in untreated infants, and we correlated the results with magnetic resonance imaging (MRI) of the brain at birth and at 6 months, as well as with the neurological outcomes at 12 months of life.

Materials and Methods: Forty-nine asphyxiated infants treated with total body TH within 6 hours of age (group A: gestational age [GA] 39⁺² ± 1⁺⁴ weeks) and 10 asphyxiated untreated infants (group B: GA 39⁺⁴ ± 1⁺³ weeks) were enrolled and studied up to 12 months of life. During the hospitalization, plasmatic measures of MBL and S100B were performed by ELISA immunoassay on samples collected at 0 to 24 hours (T1), 24 to 48 hours (T2), 48 to 72 hours of life (T3), and after 72 hours of life (T4) in both the groups; 750 ηg/mL was the cut off to discriminate MBL deficiency in neonates and 50 pg/mL was the cut off for S100B. All neonates underwent MRI at the admission and after 6 and 12 months. The brain damage was defined according to the MRI Barkovic score. Children were then observed in a 1-year multidisciplinary follow-up. Analyses were performed using the SPSS software. A p-value of  < 0.05 indicated statistical significance.

Results and Discussion: Mean serum levels of MBL and S100B in the first 24 hours of life (T1) and after 72 hours of life (T4) were similar in the two groups of infants (p = 0.31 and p = 0.15 for MBL; p = 0.99 and p = 0.41 for S100B). MBL significantly increased from T1 to T4 in group A (p = 0.002), while S100B significantly decreased (p = 0.037). In group B, the difference of MBL levels from T1 to T4 was significant (p = 0.009) but not the difference of S100B levels (p = 0.08). At 8 to 10 days of life, infants in group B showed significantly worse MRI than those of group A infants (p = 0.009). All neonates with a pathological MRI (Barkovich score of 3–4) and all deceased infants were carriers of the AA MBL2 genotype (p  <  0.001).

Conclusion: TH does not significantly influence MBL and S100B plasma levels. MBL significantly increases in all asphyxiated infants and this increase persists beyond 72 hours of life. In untreated infants, we did not observe significant worsening of parenchymal damages on MRI at 6 months. The wild-type genotype of MBL2 gene was associated to a worst short-term outcome. We observed no significant differences in 6 to 12 months MRI in the two groups of infants and in the neurological outcomes at 12 months of life.

Keywords: therapeutic hypothermia, mannose-binding lectin, mannose-binding lectine, gene 2 S100 B protein, neonatal brain damage, neurological outcomes, Barkovich score, brain MRI