Early clinical markers for the development of bronchopulmonary dysplasia: Soluble E-Selectin and ICAM-1

Objective. To test the hypothesis that in infants born at less than or equal to 29 weeks' gestation soluble adhesion molecule concentrations would be higher in the first week of life in infants that subsequently develop 'bronchopulmonary dysplasia than in infants that do not. Design and Methods. In cord blood and on days of life 1, 3, and 7, we measured serum concentrations of soluble P-Selectin, E-Selectin, and intercellular adhesion molecule-1 in samples obtained from infants less than or equal to 29 weeks' gestational age. At 1 month of age we assessed the infants' clinical courses to determine whether the infants met the criteria for the diagnosis of bronchopulmonary dysplasia (BPD) and evaluated the infants' radiographic studies to stage the level of BPD. On discharge, the duration of oxygen therapy, the requirement for home oxygen therapy, and length of hospital stay were determined. Results. Concentrations of soluble P-Selectin were greatest in cord blood samples obtained from all infants and were markedly reduced on day of life 1, regardless of the subsequent development of BPD. In serum samples obtained from cord blood and on days of life 1 and 3, soluble E-Selectin levels were higher in infants that developed BPD than in infants that did not develop BPD. In addition, the highest concentrations of soluble E-Selectin in serum samples from cord blood and on day of life 1 were associated with the development of stage 3 or 4 BPD. Soluble intercellular adhesion molecule-1 concentrations were higher on days of life 3 and 7 in the infants that went on to develop BPD than in those that did not. Conclusions. Because neutrophil attachment to endothelial cell adhesion molecules is a key event in the initiation of an inflammatory response, the association of higher early concentrations of soluble E-Selectin with the development of BPD suggests that E-Selectin may play a key role in the pathogenesis of lung inflammation and the development of BPD. This association also suggests that inflammatory events or effects leading to inflammatory responses occurring in the prenatal and/or very early perinatal periods contribute significantly to the pathogenesis of BPD.