Performance of a host-protein score (MeMed BV score) for detecting bacterial infections in young children.
In a prospective exploratory study of 69 young children with fever without source, the MeMed BV host-protein score had 47.1% sensitivity, 84.6% specificity, and an AUC of 0.80 for adjudicated bacterial infection, with lower sensitivity than procalcitonin.
Open original publication →What the AI sees
In a prospective exploratory study of 69 young children with fever without source, the MeMed BV host-protein score had 47.1% sensitivity, 84.6% specificity, and an AUC of 0.80 for adjudicated bacterial infection, with lower sensitivity than procalcitonin.
Research significance
The evidence indicates that MeMed BV, at the evaluated thresholds and classification approach, misses many bacterial infections in young children; it may therefore inform future diagnostic or antibiotic-selection research, but any value in pediatric oncology or immunocompromised children is an untested inference.
Source abstract
UNLABELLED: Diagnosing bacterial infections in children with fever without source remains challenging. The MeMed BV score, combining C-reactive protein, interferon gamma-induced protein-10, and tumor necrosis factor-related apoptosis-inducing ligand, has shown high performance in detecting bacterial infections, mostly in adults. The objective of this study is to explore the diagnostic performance of the MeMed BV score in identifying bacterial infections in young children with fever without source. In this prospective exploratory observational study, children younger than 3 years presenting to the pediatric emergency departments with acute fever without source were eligible for inclusion. An adjudication committee reviewed clinical and microbiological data to establish the diagnosis of bacterial infection. The MeMed BV score ranged from 0 to 100, with manufacturer-validated thresholds: < 35 viral, > 65 bacterial, 35-65 equivocal. In the main analysis, equivocal scores were considered bacterial. A sensitivity analysis was subsequently performed excluding patients with equivocal scores. The performance of MeMed BV score was compared to procalcitonin. Among the 384 eligible children, 69 were analyzed. Median age was 2 months (IQR 1-10), and 25% had a bacterial infection. The sensitivity of MeMed BV was 47.1% (95% CI 23.0-72.2), the specificity was 84.6% (95% CI 72.1-93.1), and the area under the curve was 0.80 (95% CI 0.69-0.92). Procalcitonin's sensitivity was 82.4% (95% CI 56.6-96.2). CONCLUSION: In infants under 3 years old with fever without source, the MeMed BV showed limited sensitivity and warrants further evaluation before routine clinical implementation. WHAT IS KNOWN: • The MeMed BV score, a diagnostic test combining three proteins, has shown good performance in adults for detecting bacterial infections. • Children under 3 years of age are particularly at risk of bacterial infections, making early diagnosis crucial. WHAT IS NEW: • In this study, MeMed BV showed low sensitivity (47.1%), missing several bacterial infections. • These findings suggest that MeMed BV should be interpreted with caution when used to identify bacterial infections in young children presenting with fever without a source.