Diagnostic accuracy of septic markers for neonatal sepsis

Abstract

Background Neonatal sepsis is a major cause of morbidity andmortality. A positive blood culture is the gold standard fordiagnosis of neonatal sepsis. The signs and symptoms suggestingneonatal sepsis are non-specific. There is no rapid and reliablelaboratory test findings for confirmation of etiologic diagnosis.Clinical signs, symptoms, and laboratory examinations are notperceived as sensitive or specific for diagnosis of sepsis.Objective The purpose of this study was to evaluate the accuracyof the septic markers for diagnosis of neonatal sepsis.Methods Blood culture was used as gold standard to compareseptic markers to diagnose neonatal sepsis. Sensitivity, specificity,positive predictive value (PPV), negative predictive value (NPV),positive and negative likelihood ratio (LR), and accuracy werecalculated.Results We identified 130 cases suspected of neonatal sepsis duringSeptember 2005 until March 2006. Four patients were excludedbecause of major congenital anomalies. The mean age was 2.2 daysand 51.6% were boys. We found fifty six (44.4%) neonates havepositive blood culture. All of septic markers had sensitivity morethan 80%. Immature to Total Neutrophil ratio (Iff) ratio had thehighest sensitivity (96.4%) and C-Reactive Protein (CRP) had thelowest sensitivity (80.4o/o). Combination among leukocyte count,thrombocyte, and Iff ratio had the highest sensitivity (sensitivitywas 85. 7%, specificity was 97.1 o/o, positive predictive value was95.9%, negative predictive value was 89.5%, accuracy was 94.4%,and positive likelihood ratio was 30.0).Conclusion Septic markers can be used in the diagnosticevaluation of neonates with suspected sepsis.