Abstract
BACKGROUND: Resistin (RETN) levels are potential diagnostic markers for sepsis in neonates and children. However, studies have yielded inconsistent results. This study aimed to compare the diagnostic accuracy of RETN levels with that of C-reactive protein (CRP) levels in the diagnosis of paediatric and neonatal sepsis through a comprehensive review of recent literature. METHODS: A standard methodology for systematic reviews and meta-analyses was followed. The PubMed, Embase and Cochrane databases were searched from January 1996 to October 2024 (PROSPERO CRD42024621872). Eligible studies were selected and analysed using Review Manager 5.4 and STATA 17. Meta-DiSc version 1.4 was used to describe and calculate the sensitivity, specificity, summary receiver operating characteristic (SROC) curves and areas under the curves (AUCs). SROC curve analysis was used to summarize the overall performance. RESULTS: A total of 437 neonates and children were included in six identified studies, all of which demonstrated reasonable methodological quality. The pooled sensitivity for the RETN level was 0.88 [95% confidence interval (CI), 0.83-0.92], which surpassed that of the CRP level at 0.85 (95% CI, 0.79-0.90). However, the pooled specificity for the RETN level was 0.78 (95% CI, 0.71-0.83), which was lower than that of the CRP level at 0.84 (95% CI, 0.77-0.90). Furthermore, the SROC curves for RETN and CRP in predicting sepsis in neonates and children indicated high predictive abilities, with AUC values of 0.925 and 0.945, respectively. CONCLUSIONS: The current evidence suggests that the RETN level is a valuable biomarker for detecting paediatric and neonatal sepsis. SYSTEMATIC REVIEW REGISTRATION: https://www.crd.york.ac.uk/PROSPERO/, identifier [CRD42024621872].