A34-05 Blood Eosinophil Dynamics in COPD: Impact of Exacerbations and Corticosteroid Use in a Nationwide Study

Background Blood eosinophil counts are pivotal for guiding advanced therapies in Chronic Obstructive Pulmonary Disease (COPD), yet their long-term consistency is poorly understood. This study aimed to characterize the longitudinal variability of blood eosinophils and identify factors associated with their instability within a large, real-world COPD population. Methods We conducted a nationwide, register-based cohort study of 63,517 Danish patients with COPD, each with at least four eosinophil measurements over a median follow-up of 3.4 years. Eosinophil levels were categorized into three groups: <0.15, 0.15-0.29, and ≥0.30 ×109/L. We used logistic regression, adjusting for multiple clinical variables including age, sex, comorbidities, and treatments, to determine predictors of stability. Patient transitions between eosinophil categories were visualized using Sankey diagrams. Findings A significant majority of patients (74%) transitioned between eosinophil categories at least once, with only 26% exhibiting a stable eosinophil profile throughout the follow-up period. While a high baseline eosinophil count (≥0.30 ×109/L) was a predictor of stability (OR 1.70, 95% CI 1.63-1.77), eosinophil counts were highly variable overall. Although severe exacerbations and treatment with oral corticosteroids or antibiotics were associated with shifts in eosinophil levels, over 80% of these shifts occurred in the absence of such clinical events. Among patients with unstable counts, 39.7% changed category once, 43.2% twice, and 17.1% three times. Interpretation In patients with COPD, blood eosinophil counts demonstrate substantial long-term variability, often independent of acute clinical events. Consequently, relying on a solitary measurement to guide therapeutic decisions may be insufficient. Serial monitoring of eosinophil counts could offer a more reliable method for identifying patients with persistent type 2 inflammation who are most likely to respond to targeted therapies.