Diagnostic performance of eosinopenia for identifying bacterial infections in the emergency department: a prospective multicenter study

Bacterial infections remain a major cause of morbidity and mortality worldwide. In the Emergency Department (ED), timely diagnosis and initiation of appropriate antibiotics are crucial to improve patient outcomes. While leukocyte and neutrophil counts are routinely interpreted, eosinophils are rarely considered, despite being systematically available. We aimed to assess the discriminative performance of eosinopenia compared with standard leukocyte parameters for identifying bacterial infections in the ED. We conducted a prospective, multicenter study in five French EDs between March 2018 and September 2021. Adult patients with suspected or confirmed bacterial infection who underwent a differential blood count were included, alongside a control group of non-infected ED patients. Eosinophil counts were compared with leukocytes, neutrophils, and C-reactive protein (CRP). The diagnosis of infection was confirmed during hospital follow-up by independent expert review. A total of 770 patients were analyzed, including 422 infected and 348 controls. Median eosinophil count was significantly lower in patients with bacterial infection than in controls (10/mm3 vs 80/mm3, p<0.001). Eosinopenia ≤10/mm3 had a specificity of 88% and a sensitivity of 48%. The discriminative performance, evaluated by the area under the curve (AUC), of eosinopenia (AUC 0.75) was comparable to neutrophils (AUC=0.74) and leukocytes (AUC=0.72), though inferior to CRP (AUC=0.89). Across infection sites, eosinopenia consistently ranked as the best-performing leukocyte marker, with maximal performance in urinary infections (AUC=0.82). Eosinophil count was not associated with in-hospital mortality in infected patients (HR=1.001, 95%CI: [0.997-1.005]). Eosinopenia represents a simple, low-cost, and readily available biomarker for diagnosing bacterial infections in the ED. Its discriminative performance is comparable to that of traditional leukocyte parameters, although it remains inferior to CRP. These findings support its role as a complementary rule-in biomarker reinforcing early clinical suspicion of infection in the ED.

Authors

Institutions

Publication Details

Journal
BMC Emergency Medicine
Published
2026-04-24
DOI
https://doi.org/10.1186/s12873-026-01581-0
Primary Topic
Neutropenia and Cancer Infections
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Diagnostic performance of eosinopenia for identifying bacterial infections in the emergency department: a prospective multicenter study

Charles-Éric Lavoignet, Pascale Dussert, Laure Abensur Vuillaume, Pascal Bilbault et al.
BMC Emergency Medicine
Neutropenia and Cancer Infections
article

Diagnostic performance of eosinopenia for identifying bacterial infections in the emergency department: a prospective multicenter study

Charles-Éric Lavoignet, Pascale Dussert, Laure Abensur Vuillaume, Pascal Bilbault, Julia Chevrolet-Lavoignet, Pierrick Le Borgne, Fanny Schweitzer, Suheil Zayet, Valentin De Carlo, François Lefebvre, Laura Bejinariu
article en

Abstract

Bacterial infections remain a major cause of morbidity and mortality worldwide. In the Emergency Department (ED), timely diagnosis and initiation of appropriate antibiotics are crucial to improve patient outcomes. While leukocyte and neutrophil counts are routinely interpreted, eosinophils are rarely considered, despite being systematically available. We aimed to assess the discriminative performance of eosinopenia compared with standard leukocyte parameters for identifying bacterial infections in the ED. We conducted a prospective, multicenter study in five French EDs between March 2018 and September 2021. Adult patients with suspected or confirmed bacterial infection who underwent a differential blood count were included, alongside a control group of non-infected ED patients. Eosinophil counts were compared with leukocytes, neutrophils, and C-reactive protein (CRP). The diagnosis of infection was confirmed during hospital follow-up by independent expert review. A total of 770 patients were analyzed, including 422 infected and 348 controls. Median eosinophil count was significantly lower in patients with bacterial infection than in controls (10/mm3 vs 80/mm3, p<0.001). Eosinopenia ≤10/mm3 had a specificity of 88% and a sensitivity of 48%. The discriminative performance, evaluated by the area under the curve (AUC), of eosinopenia (AUC 0.75) was comparable to neutrophils (AUC=0.74) and leukocytes (AUC=0.72), though inferior to CRP (AUC=0.89). Across infection sites, eosinopenia consistently ranked as the best-performing leukocyte marker, with maximal performance in urinary infections (AUC=0.82). Eosinophil count was not associated with in-hospital mortality in infected patients (HR=1.001, 95%CI: [0.997-1.005]). Eosinopenia represents a simple, low-cost, and readily available biomarker for diagnosing bacterial infections in the ED. Its discriminative performance is comparable to that of traditional leukocyte parameters, although it remains inferior to CRP. These findings support its role as a complementary rule-in biomarker reinforcing early clinical suspicion of infection in the ED.

BMC Emergency Medicine
Inserm (FR), Regenerative NanoMedicine (FR), Bruker (France) (FR), Centre Hospitalier Régional de Metz-Thionville (FR), Instituto Florestal (BR), L'Hôpital Nord Franche-Comté (FR), Hôpitaux Universitaires de Strasbourg (FR), Groupe Hospitalier de la Région de Mulhouse et Sud Alsace (FR), Université de Strasbourg (FR)
Reduced inequalities
Openalex Percentile: Top 14%
Neutropenia and Cancer Infections
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.