Our findings indicate that copeptin cord blood concentrations reflect perinatal stress with the highest values found in neonates with asphyxia. To the best of our knowledge, this is the first study to investigate copeptin in newborns with EOS and asphyxia.
Copeptin cord blood concentration was strongly and inversely correlated with umbilical artery pH, umbilical artery base excess, and with pH and lactate at admission to the NICU. Perinatal acidosis results from diminished fetal blood and oxygen supply due to maternal, placental or cord complications leading to lactic acidosis. Vasopressin is released by the hypothalamus-hypopituitary upon sensing of increased plasma osmolality, decreased arterial pressure, and reductions in cardiac volume . Our data indicate that the vasopressin system in the neonate is strongly activated upon perinatal stress. Importantly, the strength of these correlations was comparable between very preterm, late preterm and term infants, suggesting that the vasopressin response is already present at an early gestational age.
The highest copeptin cord blood concentrations were found in neonates with perinatal asphyxia. This finding was confirmed by multivariate analysis adjusted for gestational age, birth weight, mode of delivery and umbilical artery pH. Most values in this group, a third of them exceeding 2000 pmol/l, were much higher that concentrations that have been reported in adult studies on patients with septic shock, multiple trauma or myocardial infarction [9, 25, 26]. Copeptin cord blood concentrations above 400 pmol/l had a high sensitivity and specifity for asphyxia. Perinatal asphyxia can be considered as an extreme of a stress situation. In a study on the stress response to hypoxia in neonatal piglets, maintenance of cardiovascular function and a higher serum cortisol concentration were associated with a better neurological outcome . Further studies are needed to determine whether copeptin is related to asphyxia severity and whether copeptin may improve outcome prognostication after asphyxia.
Neonates born by vaginal delivery had significantly elevated copeptin cord blood concentrations compared to those born by cesarean section, even after adjustment for gestational age. Wellmann et al. have recently determined copeptin in 177 neonates and found higher levels after vaginal delivery as well . These findings are in line with earlier reports of elevations in the stress hormone cortisol after vaginal delivery . Spontaneous labour physiologically induces important changes in the fetal homeostatic system which serve to prime the fetus for postnatal adaptation.
Copeptin has been shown to be a valuable infection marker in adults with sepsis and community-acquired pneumonia [9, 12–14]. In contrast, in our study, copeptin concentrations in cord blood were not significantly elevated in EOS infants compared to controls. The specifity and sensitivity of copeptin in the diagnosis of EOS was poor. Similarly, no difference was found between neonates born to mothers with chorioamnionitis and controls. Given the strong influence of perinatal stress on copeptin cord blood concentrations, and considering the very large interindividual variations observed in this study, our data indicate that determining copeptin concentrations is not suitable to diagnose EOS. Potentially, the diagnostic accuracy of copeptin may be improved in late-onset infections, since the effect of perinatal stress on copeptin disappears over the first days of life .
Several limitations of this study need to be mentioned. Firstly, only neonates where cord blood was available were included. A selection bias is, however, unlikely, since cord blood was routinely collected during the study period in neonates of mothers with unknown or negative Toxoplasma gondii serostatus, a condition which is unlikely to affect copeptin cord blood concentrations. Secondly, the relatively small sample sizes limits statistical power. Therefore, confirmation by future prospective cohorts is needed.
We believe that the present study has several strengths. In contrast to the study by Wellmann et al.  which included only healthy term and near term infants, we included neonates with a wide range of gestational ages. The inclusion of clearly defined and not overlapping groups of infants with EOS, chorioamnionitis and asphyxia allowed to study the influence of these diseases on copeptin concentration. Multivariate analyses were adjusted for the main confounders gestational age, birth weight, delivery mode and umbilical artery pH.