RESUMO
BACKGROUND: Urinary tract infections (UTIs) are very common bacterial infections in children. Early detection of renal parenchymal involvement in this setting can help clinicians make more effective treatment choices. The aim of this pilot study was to assess the ability of plasma and urinary neutrophil gelatinase-associated lipocalin (pNGAL and uNGAL) levels, measured using an automated system, to accurately predict renal parenchymal involvement in children with febrile UTIs. METHODS: This prospective single-center study included 28 children aged ≥ 4 years with a first episode of febrile UTIs. All patients underwent magnetic resonance imaging. pNGAL, uNGAL, procalcitonin, C-reactive protein (CRP), and white blood cells were measured before antibiotic therapy. RESULTS: The receiver operating characteristic (ROC) area under the curve for predicting acute pyelonephritis was 0.6 for pNGAL, 0.8 for CRP, 0.4 for PCT, and 0.4 for uNGAL. The ROC analyses showed an optimal cutoff of 141.0 ng/mL for pNGAL (sensitivity, 54.2%; specificity, 75.0%; positive predictive value, 92.9%; and negative predictive value, 21.4%). CONCLUSION: pNGAL and uNGAL did not effectively aid the early prediction of renal parenchymal involvement in children ≥ 4 years with febrile UTIs. The novelties of this study were the use of MRI as the gold standard and an automated biochemical method to measure NGAL.
RESUMO
Introduction: A reliable pediatric triage tool is essential for nurses working in pediatric emergency departments to quickly identify children requiring priority care (high-level emergencies) and those who can wait (low-level emergencies). In the absence of a gold standard in France, the objective of our study was to validate our 5-level pediatric triage tool -pediaTRI- against the reference tool: the Pediatric Early Warning Score (PEWS) System. Materials and Methods: We prospectively included 100,506 children who visited the Pediatric Emergency Department at Lenval Children's Hospital (Nice, France) in 2016 and 2017. The performance of pediaTRI to identify high-level emergencies (severity levels 1 and 2) was evaluated in comparison with a PEWS ≥ 4/9. Data from 2018-19 was used as an independent validation cohort. Results: pediaTRI agreed with the PEWS score for 84,896 of the patients (84.5%): 15.0% (14.8-15.2) of the patients were over-triaged and 0.5% (0.5-0.6) under-triaged compared with the PEWS score. pediaTRI had a sensitivity of 76.4% (74.6-78.2), a specificity of 84.7% (84.4-84.9), and positive and negative likelihood ratios of 5.0 (4.8-5.1) and 0.3 (0.3-0.3), respectively, for the identification of high-level emergencies. However, the positive likelihood ratios were lower for patients presenting with a medical complaint [4.1 (4.0-4.2) v 10.4 (7.9-13.7 for trauma), and for younger children [1.2 (1.1-1.2) from 0 to 28 days, and 1.9 (1.8-2.0) from 28 days to 3 months]. Conclusion: pediaTRI has a moderate to good validity to triage children in a Pediatric Emergency Department with a tendency to over-triage compared with the PEWS system. Its validity is lower for younger children and for children consulting for a medical complaint.