Early-onset sepsis remains a major diagnostic problem in neonatal medicine. Definitive diagnosis depends on cultures of blood or other normally sterile body fluids. Abnormal hematological counts, acute-phase reactants, and inflammatory cytokines are neither sensitive nor specific, especially at the onset of illness. Combinations of measurements improve diagnostic test performance, but the optimal selection of analytes has not been determined. The best-established use of these laboratory tests is for retrospective determination that an infant was not infected, based on failure to mount an acute-phase response over the following 24 to 48 hours.