Early diagnosis of neonatal infection has proved problematic due to the inadequacy of currently available laboratory tests. Neonatal sepsis is associated with an increase in plasma-derived cytokine levels, but an increase of a single cytokine cannot identify neonatal sepsis specifically and multiple cytokine levels are required. The time constraints and relatively large volume of plasma required to measure multiple cytokines from newborn infants by conventional enzyme-linked immunosorbent assay (ELISA) techniques is prohibitive. We therefore applied cytometric bead array (CBA) technology for simultaneous measurement of multiple cytokines from a group of 18 term neonates with infection confirmed by culture and a control group. 'Normal' ranges were established for each cytokine from 1-7-, 8-14- and 15-21-day-old newborns. There was no significant change in the levels of cytokines from infants in different control age groups, suggesting that basal cytokine levels are unchanged in the first 3 weeks of life. In the patient groups, however, there was a significant difference in several cytokines between the different age groups. Interleukin (IL)-6, IL-10 and IL-12 were increased significantly in the 1-7-day-old patient group compared to either the 8-14 and 15-21 age group, suggesting that infection in utero is associated with increased levels of these cytokines compared to infection acquired following birth. When individual patient cytokine levels were compared to normal control reference ranges, two patients failed to show significant elevation of any cytokine tested. All other patients showed elevated levels of between one and nine cytokines tested (mean of 4.6). There was no correlation between elevated cytokine levels and types of infective organism or patient age. In conclusion, neonatal sepsis is associated with the elevation of multiple plasma cytokines. The use of CBA kits is a rapid, easy, low sample volume and sensitive method to measure multiple plasma cytokines.