Diagnostic accuracy of interleukin-6 for early-onset sepsis in preterm neonates.

Early-onset sepsis (EOS) is a leading cause of morbidity and mortality among neonates. Yet, accurate diagnosis remains a major challenge in clinical routine. The aim of this study was to evaluate the diagnostic accuracy of Interleukin-6 (IL-6) in combination with other objective perinatal data for early-onset sepsis (EOS) in preterm neonates. We conducted a retrospective nested case-control study with preterm neonates with a birth weight < 2000 g born in our NICU between January 2007 and June 2016. Differences of IL-6 levels and other perinatal clinical and laboratory data between neonates with and without EOS were statistically analyzed. Sixty-seven preterm infants with and 115 neonates without EOS were included in this study. Specificity and sensitivity for IL-6 were 72.8% and 75.0%, respectively, with an area under the curve of 0.804 at a cut-off point of 40 ng/l. Depending on the statistical method applied, combining IL-6 with a second perinatal factor led either to an increase of specificity (82.4-100%) or sensitivity (75.0-92.2%). The combination of IL-6 with other perinatal factors can significantly increase specificity and sensitivity in the diagnosis of EOS. However, overall diagnostic accuracy cannot be notably improved as there is a tradeoff between sensitivity and specificity. Although these findings do not necessarily apply in clinical routine, they can be of substantial value in the assistance of individual decision making.