Blood eosinophil count and intensive care unit admission in patients hospitalized for acute exacerbation of COPD: a retrospective cohort study.

backgroundAcute exacerbations of chronic obstructive pulmonary disease (COPD) are major clinical events associated with hospitalization, intensive care unit (ICU) admission, and increased mortality. Blood eosinophil count has emerged as a potential inflammatory biomarker in COPD; however, its role in predicting ICU requirement among hospitalized patients remains unclear.

methodsThis retrospective cohort study included patients aged ≥ 40 years hospitalized with acute COPD exacerbation at Ankara Bilkent City Hospital between 2020 and 2024. COPD diagnosis was based on Global Initiative for Chronic Obstructive Lung Disease (GOLD) criteria. Blood eosinophil counts measured within the first 60 min of presentation to the emergency department were categorized as < 150 cells/µL, 150-299 cells/µL, and ≥ 300 cells/µL. ICU admission was defined as the primary outcome. Competing risk analysis using the Fine-Gray model was performed, with death as a competing event.

resultsA total of 385 patients were included. Eosinophil groups were not significantly associated with ICU requirement. Clinical variables significantly associated with ICU admission included GOLD group, GOLD grade, modified Medical Research Council (mMRC) dyspnea score, long-term oxygen therapy (LTOT), nebulizer and non-invasive mechanical ventilation (NIMV) use, and need for NIMV at admission. In Fine-Gray models, GOLD group E was independently associated with increased ICU admission risk compared with GOLD A (subdistribution hazard ratio [sHR] = 3.35, 95% confidence interval [CI] = 1.57-7.15; p = 0.002), whereas eosinophil levels were not.

conclusionIn hospitalized patients with acute COPD exacerbations, blood eosinophil count - regardless of the threshold applied - was not independently associated with ICU admission after adjustment for disease severity and competing risks. GOLD group E emerged as the strongest independent predictor of ICU requirement in the same model. These findings suggest that clinical severity assessment, rather than eosinophil-based stratification, should guide risk assessment for ICU admission in this population. CLINICAL TRIAL NUMBER: Not applicable.