Admission Absolute Eosinophil Count can Discriminate Influenza from Other Respiratory Pathogens in Hospitalized Children with Lower Respiratory Tract Infection
Year: 2026
Type: preprint
Study design: Observational Study
Abstract: Background Given that influenza is frequently diagnosed late in children when nucleic acid testing is unavailable, a low-cost indicator obtained from routine blood counts would be clinically useful. Therefore, we evaluated the absolute eosinophil count (EOS) in children with and without influenza. Methods This retrospective cohort study enrolled 252 hospitalized children with lower respiratory tract infections (122 with influenza and 130 without influenza). A multivariable linear regression was fitted with ln(EOS) as the dependent variable and influenza status as the exposure variable, adjusted for age, sex, and glucocorticoid exposure. Discriminant performance was assessed against routine hematological indices, high-sensitivity C-reactive protein (hs-CRP), and interleukin-6, using analyses including the area under the receiver operating characteristic curve (AUC). A paired subgroup of 92 children underwent repeat blood counts before discharge. Results Admission EOS in the influenza group (0.0175×10⁹/L) was significantly lower than that in the non-influenza group (0.1565×10⁹/L; P < 0.001). EOS below the reportable limit was detected in 76.2% of cases (adjusted odds ratio (OR) 6.61, 95% confidence interval (CI) 3.54–12.34), and median EOS was lower than that in all other groups (all P < 0.001). In the paired subgroup (44 influenza, 48 non-influenza), EOS increased from 0.0215 to 0.0745 ×10⁹/L in children with influenza ( P < 0.001) and from 0.0869 to 0.1363 ×10⁹/L in children without influenza ( P = 0.087); the subgroup was underpowered to compare the magnitude of change between groups. Conclusions Influenza was associated with markedly lower admission EOS, which increased during recovery. EOS added discriminative information beyond routine hematological indices and hs-CRP, and its high negative predictive value supports EOS as a low-cost adjunct for frontline influenza recognition but not as a standalone trigger for antiviral therapy.
Source: Research Square
Language: English
Subfield: Epidemiology
Field: Medicine
Domain: Health Sciences
Open Access status: green
OpenAlex ID: W7225147165