C-reactive Protein as a Screening Biomarker in Neonatal Sepsis

OBJECTIVE: To measure validity of C-reactive protein (CRP) as screening test for neonatal sepsis (NS) and compare its screening validity between early onset neonatal sepsis (EONS) and late onset neonatal sepsis (LONS). STUDY DESIGN: Cross-sectional study Place and Duration of Study: Neonatal Unit, Town Children Hospital, Peshawar, Khyber Pukhtunkhawa, from August 2016 to February 2017. METHODOLOGY: A total of 385 neonates from age 0 to 28 days with clinical features of neonatal sepsis were sampled using consecutive sampling technique. Two groups were identified, i.e. early onset neonatal sepsis (age <72 hours) and late onset neonatal sepsis (age >72 hours). Each neonate was sampled for blood culture and C-reactive protein (CRP). Diagnosis of neonatal sepsis was established through a positive blood culture. Data was analysed using SPSS V 25.0. Sensitivity, specificity, negative predictive value (NPV) and positive predictive value (PPV) of CRP was measured and compared in each group. RESULTS: Analysis showed a low validity of CRP as screening test in neonatal sepsis (Sensitivity=35.525%, specificity=58.0%, PPV=85% and NPV=11.83%). Initial screening test validity of CRP was low in EONS (sensitivity=17.16%, specificity=58.33%, PPV=72.72% and NPV=9.81%) compared to LONS (sensitivity=77.45%, specificity=57.14%, PPV=92.94% and NPV=25.80) Conclusion: CRP as a screening test has low screening validity in early onset neonatal sepsis compared to late onset sepsis.