Blood eosinophil counts during severe exacerbations do not reflect stable state inflammatory phenotype in COPD

Abstract

<b>Introduction:</b> GOLD guidelines endorse use of blood eosinophil counts (BEC) to guide inhaled corticosteroid (ICS) treatment in COPD, but do not stipulate sampling during clinical stability. BEC may fall during acute illness. In severe COPD exacerbations, admission BEC &lt;0.05 is a strong adverse prognostic index¹. <b>Aims:</b> The blood eosinophils in COPD (BECCOPD) study is assessing whether the highest of ≥3 BEC within the past 24 months (historical BEC) is a suitable surrogate for stable state BEC. We assessed the impact of severe exacerbation on BEC and the influence on ICS prescribing. <b>Methods:</b> Eligible patients, stable at study entry, were recruited from primary and secondary care. In a subgroup hospitalised for COPD exacerbation within the study window, we compared the agreement between study entry BEC and admission, pre-discharge, stable and historical BEC. <b>Results:</b> Of 146 BECCOPD patients, 53 were admitted for COPD exacerbation. Mean (SD) age 68 (8), mean FEV1 49% (21%) predicted. Median (IQR) eMRCD score 4 (3-4), 51% were male. Agreement between BEC samples is tabulated. <b>Conclusion:</b> BEC during hospitalisation for COPD exacerbation are not a reliable estimate of stable-stable eosinophil phenotype and should not be used to guide ICS therapy. Historical BEC offer good performance at the 0.1 threshold (1/53 ICS potentially incorrectly withheld), allowing timely optimisation of therapy within a high-risk period. <b>