Predictability of biomarkers in diagnosis of neonatal sepsis: a cross sectional study

Abstract

Background: There is lack of sensitive and specific markers for diagnosis of early neonatal sepsis. Procalcitonin and CRP are known sensitive marker of sepsis and CRP is known to be affected by other independent variables. Hence this study was conducted to compare their predictability for diagnosis of early neonatal sepsis and to assess the effect of variables on them. Methods: It is a cross sectional study. Study group comprised of neonates with definitive (n=09) and probable (n=24) signs, symptoms and laboratory markers of sepsis compared with no sepsis (n=67). Procalcitonin and CRP were measured by immunofluoroscence and immunoturbidometric methods respectively in serum of neonates admitted to intensive care unit on their first day of admission. Fishers test, median test, ROC curve and multiple regressions were used for statistical analysis. Results: Procalcitonin (p=0.007) and CRP (p<0.001) levels were high in sepsis. Procalcitonin was not significantly affected by independent variables in asymptomatic neonates (r 2 =0.12). ROC curve analysis revealed the predictability of procalcitonin (area under curve - 0.92) and CRP (area under curve-0.74) for the diagnosis of neonatal sepsis. Conclusions: Procalcitonin found to have better predictability than CRP in the diagnosis of neonatal sepsis.