Lipopolysaccharide-binding protein, lipopolysaccharide, and soluble CD14 in sepsis of critically ill neonates and children.

OBJECTIVE: To compare the diagnostic accuracy of lipopolysaccharide-binding protein (LBP) for sepsis in critically ill neonates and children with the two markers participating in the same inflammatory pathway, lipopolysaccharide and soluble CD14.

DESIGN AND SETTING: Prospective, observational study in a multidisciplinary neonatal and pediatric intensive care unit.

PATIENTS: 47 critically ill neonates and 49 critically ill children with systemic inflammatory response syndrome (SIRS) and suspected sepsis, classified into two groups: those with and those without sepsis.

INTERVENTIONS: Serum LBP, lipopolysaccharide, soluble CD14, C-reactive protein, and procalcitonin were measured on 2 consecutive days. The area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and predictive values were evaluated.

RESULTS: AUC for LBP on the first day of suspected infection was 0.97 in neonates aged under 48 h, 0.93 in neonates over 48 h and 0.82 in children. AUCs for lipopolysaccharide and soluble CD14 were 0.77 and 0.74 in neonates under 48 h, 0.53 and 0.76 in neonates over 48 h, and 0.72 and 0.53 in children. AUCs for procalcitonin and C-reactive protein were 0.65 and 0.89 in neonates under 48 h, 0.65 and 0.91 in neonates over 48 h, and 0.76 and 0.69 in children.

CONCLUSIONS: In critically ill neonates and children LBP concentration on the first day of suspected sepsis is a better marker of sepsis than lipopolysaccharide, soluble CD14, procalcitonin, and in neonates younger than 48 h and children, also a better marker than C-reactive protein. Lipopolysaccharide and soluble CD14 are not suitable markers for the differentiation of infectious and noninfectious SIRS.