Management of febrile infants ≤90 days remains one of the most critical scenarios in pediatric emergency medicine. Current clinical algorithms prioritize bacterial risk stratification (serious/invasive bacterial infections, SBI/IBI), while viral epidemiology is inconsistently integrated, despite viral pathogens accounting for the majority of febrile episodes. This doctoral research is grounded on the hypothesis that integrating three dimensions — host immune response, timing from fever onset, and viral epidemiology — can improve diagnostic accuracy over the still-dominant static, bacterium-centered paradigm. The research program was structured into four components: evaluation of the timing-dependent performance of conventional inflammatory biomarkers; retrospective epidemiological and virological analyses; prospective validation of an interferon signature as a host-response marker; and development/validation of CD64 and CD169 as translational biomarkers. The core study — a single-center, prospective investigation conducted at the Pediatric Emergency Department of IRCCS Istituto Giannina Gaslini — enrolled 61 infants (29 non-BI, 12 BI, 20 healthy controls), analyzing neutrophil CD64 and monocyte CD169 expression by flow cytometry. The CD64/CD169 ratio showed the best diagnostic performance in discriminating bacterial from non-bacterial infections, with an AUC of 0.989 (sensitivity 100%, specificity 96.6%, cut-off ≥1.27), outperforming individual biomarkers and conventional markers (PCT AUC ~0.94, CRP AUC ~0.90). The non-BI group displayed a CD169-predominant profile and the BI group a CD64-predominant profile, consistent with the opposite biological direction of host response to viral versus bacterial stimuli. Overall, findings indicate that the current bacteria-centered approach is safe but lacks specificity; viral infections should occupy a central role in the diagnostic paradigm, being etiologically predominant and capable of modulating — without abolishing — bacterial risk; and host-response biomarkers, particularly the CD64/CD169 combination, represent a promising, biologically grounded path toward a dynamic diagnostic framework, potentially reducing invasive procedures, inappropriate antibiotic use, and unnecessary hospitalizations, pending confirmation in larger cohorts.
CD64 and CD169 as Translational Biomarkers for Discriminating Bacterial from Viral Infections in Febrile Infants ≤90 Days of Age: From Immune Pathophysiology to Bedside Application
BELLINI, TOMMASO
2026-07-17
Abstract
Management of febrile infants ≤90 days remains one of the most critical scenarios in pediatric emergency medicine. Current clinical algorithms prioritize bacterial risk stratification (serious/invasive bacterial infections, SBI/IBI), while viral epidemiology is inconsistently integrated, despite viral pathogens accounting for the majority of febrile episodes. This doctoral research is grounded on the hypothesis that integrating three dimensions — host immune response, timing from fever onset, and viral epidemiology — can improve diagnostic accuracy over the still-dominant static, bacterium-centered paradigm. The research program was structured into four components: evaluation of the timing-dependent performance of conventional inflammatory biomarkers; retrospective epidemiological and virological analyses; prospective validation of an interferon signature as a host-response marker; and development/validation of CD64 and CD169 as translational biomarkers. The core study — a single-center, prospective investigation conducted at the Pediatric Emergency Department of IRCCS Istituto Giannina Gaslini — enrolled 61 infants (29 non-BI, 12 BI, 20 healthy controls), analyzing neutrophil CD64 and monocyte CD169 expression by flow cytometry. The CD64/CD169 ratio showed the best diagnostic performance in discriminating bacterial from non-bacterial infections, with an AUC of 0.989 (sensitivity 100%, specificity 96.6%, cut-off ≥1.27), outperforming individual biomarkers and conventional markers (PCT AUC ~0.94, CRP AUC ~0.90). The non-BI group displayed a CD169-predominant profile and the BI group a CD64-predominant profile, consistent with the opposite biological direction of host response to viral versus bacterial stimuli. Overall, findings indicate that the current bacteria-centered approach is safe but lacks specificity; viral infections should occupy a central role in the diagnostic paradigm, being etiologically predominant and capable of modulating — without abolishing — bacterial risk; and host-response biomarkers, particularly the CD64/CD169 combination, represent a promising, biologically grounded path toward a dynamic diagnostic framework, potentially reducing invasive procedures, inappropriate antibiotic use, and unnecessary hospitalizations, pending confirmation in larger cohorts.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.



