Identifying early surrogate markers for blood culture positive sepsis in an urban NICU-A retrospective observational study

Authors : Arnab Sengupta, Vishnu Tej, Rupesh Bansal, N Saravanan, Uma Raju, Akshay Reddy, M Padmanabh Reddy

DOI : 10.18231/j.ijmpo.2021.007

Volume : 7

Issue : 1

Year : 2021

Page No : 28-35

Background: Neonatal sepsis is a major problem worldwide. Positive blood culture, the diagnostic gold standard is expensive, time consuming and yields result in limited cases. Hence reliable surrogate markers are needed.
Aims and Objective: To ascertain haematological parameters which are reliable adjuvants to blood culture for diagnosing Neonatal Sepsis.
Materials and Methods: A retrospective observational study was carried over two years in alevel3 NICU among neonates admitted with suspected sepsis. Data on demographics, clinical background, blood laboratory results were extracted from medical records and statistically analysed by multivariate regression analysis to identify surrogate markers for sepsis.
Results: Over two years, 182 neonates admitted with Suspected Sepsis were enrolled in the study. In all clinical examination followed by laboratory investigations viz CBP (TLC, ANC, ITR, degenerative changes, platelet count), CRP and blood culture was carried out prior to treatment. A total of 64 (35.2%) were blood culture positive. On multivariable regression analysis, ITR >0.2 [OR 16.3; 95% CI (1.5-176.8), platelet count < 150> Conclusions: The study revealed that Immature to total neutrophil ratio >0.2, platelet count< 150> Key Messages: 1. Blood culture, the gold standard for diagnosing neonatal sepsis is expensive, time consuming and yields positive results in limited cases. 2. Effective surrogate markers are Immature to total neutrophil ratio (ITR), thrombocytopenia and cytotoxic degenerative changes in neutrophils. 3. These surrogate markers would identify the neonate at risk requiring early antibiotic therapy.

Keywords: ANC, Blood Culture, ITR, Neonatal Sepsis, Surrogate Markers.


Citation Data


Related Articles