Circulating chemokine levels in febrile infants with serious bacterial infections.

Early diagnosis of serious bacterial infections (SBI) in febrile young infants based on clinical symptoms and signs is difficult. This study aimed to evaluate the diagnostic values of circulating chemokines and C-reactive protein (CRP) levels in febrile young infants < 3 months of age with suspected SBI. We enrolled 43 febrile young infants < 3 months of age with clinically suspected SBI who were admitted to the neonatal intensive care unit or complete nursing unit of the pediatric department of Kaohsiung Medical University Hospital between December 2006 and July 2007. Blood was drawn from the patients at admission, and complete blood counts, plasma levels of CRP, granulocyte colony-stimulating factor (G-CSF), and chemokines, including interleukin-8 (IL-8), macrophage inflammatory protein-1alpha, macrophage inflammatory protein-1beta, monokine induced by interferon-gamma, and monocyte chemotactic protein-1 were measured. Patients symptoms and signs, length of hospital stay, main diagnosis, and results of routine blood tests and microbiological culture results were recorded. Twenty-six infants (60.5%) were diagnosed with SBI, while 17 (39.5%) had no evidence of SBI based on the results of bacterial cultures. CRP, IL-8 and G-CSF levels were significantly higher in the infants with SBI than in those without SBI Plasma levels of other chemokines were not significantly different between the groups. The area under the receiver-operating characteristic (ROC) curve for differentiating between the presence and absence of SBI was 0.79 for CRP level. Diagnostic accuracy was further improved by combining CRP and IL-8, when the area under the ROC curve increased to 0.91. CRP levels were superior to IL-8 and G-CSF levels for predicting SBI in febrile infants at initial survey. IL-8 levels could be used as an additional diagnostic tool in the initial evaluation of febrile young infants, allowing clinicians to treat these patients more appropriately.