Systemic immune inflammation index and neutrophil-to-lymphocyte ratio as predictors of neonatal sepsis in low birth weight infants

Abstract Background The systemic immune inflammation index (SII) and neutrophil-to-lymphocyte ratio (NLR) are hematological parameters calculable from standard laboratory assessment. Despite extensive adult medicine implementation, neonatal applicability remains incompletely characterized. Early identification of sepsis in neonates remains challenging and sepsis is a significant global health problem associated with high morbidity and mortality, particularly among vulnerable low birth weight (LBW) infants. This study aimed to evaluate the predictive value of SII and NLR for neonatal sepsis in LBW infants. Methods This observational analytic case–control study was conducted at Dr. Hasan Sadikin General Hospital, Bandung, between January 2024 and June 2025. A total of 290 LBW infants were enrolled in this study, comprising 145 infants with neonatal sepsis and 145 without sepsis. SII and NLR values were calculated from complete blood counts obtained at the initial evaluation within six hours after birth. Data were analyzed using receiver operating characteristic (ROC) curves to determine the area under the curve (AUC), optimal cut-off values, sensitivity, specificity and likelihood ratios (LR). Results Both biomarkers were significantly associated with the subsequent development of culture-confirmed neonatal sepsis. Comparative predictive performance analysis revealed superior performance of SII (AUC: 0.778; 95%CI: 0.726–0.825) relative to NLR (AUC: 0.678; 95%CI: 0.620–0.731; p < 0.001). At an optimal cut-off value of ≤ 657,962.56, SII yielded a sensitivity of 57.9% and specificity of 88.3% with positive predictive value 83.2%, LR + 4.95 and LR − 0.48. The optimal NLR cut-off value of ≤ 3.1 showed a sensitivity of 59.3% and specificity of 73.1% with LR + of 2.20. Conclusions Both SII and NLR are significant predictors of neonatal sepsis in LBW infants. SII demonstrates superior predictive performance and higher specificity compared with NLR.