Association Between Admission Eosinophil Count and Hospital Resource Use in Community-Acquired Pneumonia

BACKGROUND: Eosinopenia has been associated with adverse outcomes in community-acquired pneumonia (CAP). However, its relationship with hospital resource use remains unclear. RESEARCH QUESTION: What is the association between admission eosinophil counts and hospital resource use among adults with CAP? STUDY DESIGN AND METHODS: This prospective multicenter cohort study (Community-Acquired Pneumonia Network of Competence [CAPNETZ]; Identifier: 2024-07-11-CHV6) has enrolled patients ≥ 18 years of age with CAP in university hospitals in Germany since 2017. Associations between admission blood eosinophil counts and hospital resource use-ICU admission, mechanical ventilation, and length of stay-were assessed using multivariable regression models. The optimal eosinophil count threshold for stratifying patients by ICU admission and mechanical ventilation rates was identified, and outcomes were compared between patients above and below this threshold. RESULTS: Lower eosinophil counts at admission were associated with increased ICU admission (n = 1,639; P < .001), including among patients treated with systemic glucocorticoids (P = .002) and those not receiving glucocorticoids (P = .047). Lower eosinophil counts also were associated with higher rates of mechanical ventilation (P = .014) and longer hospital stays (P = .024). An eosinophil count threshold of 10 cells/μL was identified as the cutoff that best distinguished patients with higher vs lower risk of ICU admission and mechanical ventilation. Patients with eosinopenia (≤ 10 cells/μL) showed higher ICU admission rates (14.2% vs 8.5%; P < .001; adjusted OR, 1.78), increased mechanical ventilation rates (9.1% vs 5.2%; P = .003; adjusted OR, 1.82), and longer hospitalization (mean, 10.2 days vs 9.0 days; P = .013). INTERPRETATION: Our results show that admission eosinopenia (≤ 10 cells/μL) was associated with greater hospital resource use and may serve as a practical biomarker for health care resource planning. CLINICAL TRIAL REGISTRATION: German Clinical Trials Register; No.: DRKS00005274; URL: https://drks.de/.