Modified hematological sepsis score in early diagnosis of neonatal sepsis

Abstract

Background: Diagnosis of neonatal sepsis at an early stage substantially reduces the mortality. The clinician often relies on laboratory parameters to support the clinical suspicion. As blood culture takes time and yield is low, hematological and biochemical parameters often guide to the diagnosis and management. Rodwell’s Hematological sepsis score (HSS) has a reasonable sensitivity but low specificity. Some of the parameters included in that scoring system are repetitive of same pathogenic mechanism. A modified HSS was developed by the authors by removing the repetitive parameters, increasing the weightage for low neutrophil count and adding a new parameter - nucleated RBC. Objective of the study was to compare the diagnostic ability of the modified hematological sepsis score with Rodwell’s hematological sepsis score.Methods: Prospective analytical study conducted in a tertiary level hospital. Neonates admitted to NICU and had complete blood count done were included. Babies with clinical signs of systemic inflammatory response syndrome and evidence of organ dysfunction were considered septic. They were classified as proven sepsis if the blood culture was positive. All the samples were scored for both HSS and modified HSS. The sensitivity, specificity and other diagnostic ability tests were compared between the two scoring systems.Results: Total of 75 neonates were enrolled. 25 of them had sepsis and three had blood culture positive. At a score of 3, the sensitivity and specificity of HSS was 80 and 70% and that of Modified HSS was 84 and 82% respectively.Conclusions: Modified hematological score improves the specificity and likelihood ratios without decreasing the sensitivity in early diagnosis of neonatal sepsis.