Prediction of mortality by nSOFA Score for late-onset sepsis in very low birth weight infants

Abstract

Introduction: Sepsis is a major cause of morbidity and mortality in premature infants. Rapid diagnosis and initiation of treatment are of great importance. This study aims to evaluate the role of the Neonatal Sequential Organ Failure Assessment (nSOFA) score in predicting the causal agents and outcomes of late-onset sepsis in preterm neonates.

Methodology: In this single-center, retrospectively designed study, nSOFA scores of preterm infants born before 32 gestational weeks and weighing under 1500 g with a diagnosis of culture-proven late-onset sepsis (LOS) were compared at different timepoints in relation to mortality.

Results: Thirteen of 117 preterms included in the study died. At all the timepoints examined, the median nSOFA score was found to be higher in the mortality group (all p < 0.001). A 3.5 cutoff value of nSOFA showed the best differentiation, with AUC = 0.97 (95% CI: 0.94–1.00), 100% sensitivity, and 91.4% specificity. When nSOFA scores were compared in patients grouped as gram-positive sepsis and gram-negative sepsis, scores at T0, T6, T12, and T24 timepoints were determined to be significantly higher in the exitus group (all p < 0.008). In preterm infants born before 28 gestational weeks, mortality was predicted with the 3.5 cutoff value at T6, T12, T24, and T48 timepoints (AUC = 0.947, 0.943, 0.972, and 0.940, respectively, all p < 0.001).

Conclusions: The results showed that the nSOFA score is useful for predicting sepsis-related mortality in preterm infants and correlates with the severity of gram-negative sepsis.