Changes in CBC and derived inflammatory indices in children with respiratory tract infections caused by Flu A, Flu B, RSV, and ADV

Background Respiratory tract infections (RTIs) are a leading infectious cause of mortality among children worldwide, with viral pathogens accounting for approximately 80% of cases. Complete blood count (CBC) and derived inflammatory indices are routinely available, but their clinical utility in differentiating common viral aetiologies remains incompletely defined. This study aimed to characterise changes in CBC parameters and derived inflammatory indices and assess their clinical utility in children with RTIs attributed to influenza A (Flu A), influenza B (Flu B), respiratory syncytial virus (RSV), and adenovirus (ADV). Methods We conducted a single-centre retrospective cohort study enrolling 7,132 children with RTIs between September 1, 2025 and February 28, 2026 in Shandong, China. Patients were divided into five groups: Flu A (n=1,169), Flu B (n=361), RSV (n=1,672), ADV (n=201), and negative control (n=3,729). CBC parameters and derived inflammatory indices were compared. Results No significant difference in sex composition was observed among the five groups. Age distribution differed significantly: RSV had the lowest median age (2.33 years), Flu A and Flu B both 6.00 years, ADV 4.83 years, and negative controls 4.58 years (P<0.001). All CBC parameters and derived inflammatory indices showed significant overall differences among five groups (P<0.001). Compared with the negative control, Flu A was characterised by elevated lymphocyte count (LYM#), lymphocyte percentage (LYM%), monocyte count (MON#) and monocyte percentage (MON%), together with reduced neutrophil percentage (NEU%), neutrophil-to-lymphocyte ratio (NLR), latelet-to-lymphocyte ratio (PLR), derived neutrophil-to-lymphocyte ratio (dNLR) and systemic immune-inflammation index (SII). Flu B exhibited elevated neutrophil count (NEU#), MON#, NEU%, NLR, monocyte-to-lymphocyte ratio (MLR), dNLR, SII, systemic inflammation response index (SIRI) and aggregate index of systemic inflammation (AISI), without significant changes in LYM# or platelet count (PLT). Both RSV and ADV showed elevated NEU#, NEU%, NLR, MLR, PLR, dNLR, SII, SIRI and AISI, accompanied by reduced LYM#, LYM% and PLT; these alterations were most pronounced in ADV, particularly for MLR, PLR, dNLR and SII. White blood cell count (WBC) was elevated in Flu A, Flu B and RSV but not in ADV. SIRI and AISI were elevated in Flu B, RSV and ADV compared with the negative control group, but did not differ significantly among these three groups. Multivariable logistic regression showed that MON# was the most stable independent risk factor for both Flu A (OR = 1.801) and Flu B (OR = 1.827),but MON# cannot be used to distinguish these two influenza types; LYM# was the strongest protective factor for ADV (OR = 0.567); MLR (OR = 2.667) and dNLR (OR = 1.289) were independent risk factors for RSV, while age (OR = 0.797) and SIRI (OR = 0.737) were protective. Receiver operating characteristic (ROC) analysis showed that individual indices had limited discriminatory ability (AUC range, 0.494–0.691), whereas combined models incorporating demographic characteristics and multiple CBC-derived indices improved discrimination (AUC range, 0.612–0.712), with the best performance for RSV (AUC = 0.712) and ADV (AUC = 0.694). Conclusion CBC and derived inflammatory indices exhibit distinct alterations in children with Flu A, Flu B, RSV, and ADV infections. Combined models incorporating multiple CBC-derived indices for target viruses may facilitate early differentiation of viral, but they cannot replace pathogen-specific polymerase chain reaction (PCR) testing.