BMC Pediatr. 2026 Jun 6. doi: 10.1186/s12887-026-07103-y. Online ahead of print.
ABSTRACT
BACKGROUND: This study aimed to investigate the clinical utility of the C-reactive protein/albumin ratio (CAR) and the neutrophil/lymphocyte ratio (NLR) in evaluating the severity of sepsis in pediatric patients.
METHODS: A retrospective cohort study was conducted involving 174 pediatric patients hospitalized between June 2022 and November 2025. Sepsis and septic shock were defined according to the 2024 Phoenix criteria. Patients were stratified into a sepsis group (n = 123) and a septic shock group (n = 51). CAR and NLR were calculated from blood samples collected within 24 h of admission. A two-sided p < 0.05 was considered statistically significant. Independent associated factors for septic shock were identified using binary logistic regression analysis. Receiver operating characteristic (ROC) curves were constructed, and the area under the curve (AUC) was calculated to evaluate the diagnostic performance of CAR, NLR, and other relevant biomarkers.The study was conducted in accordance with the STROBE guidelines.
RESULTS: Univariate analysis revealed significantly elevated white blood cell (WBC) count, CAR, and NLR levels in the septic shock group compared to the sepsis group (all p < 0.05). Binary logistic regression analysis confirmed that elevated CAR (OR = 2.893, p < 0.001) and NLR (OR = 1.232, p < 0.001) were independent associated factors for progression to septic shock. ROC curve analysis demonstrated that a model combining CAR and NLR exhibited superior discriminative performance compared to either biomarker alone, with a higher AUC (0.802), sensitivity (74.50%), and specificity (74.80%).
CONCLUSION: CAR and NLR appear to be significant independent biomarkers for assessing severity and predicting septic shock in children with sepsis. The combination of CAR and NLR provides improved predictive accuracy compared with either marker alone, suggesting its potential utility in guiding early clinical decisions.
PMID:42251342 | DOI:10.1186/s12887-026-07103-y
