Blood Eosinophil Count Predicts Treatment Failure and Hospital Readmission for COPD

We examined associations between blood eosinophil counts (BEC) and risk of treatment failure or hospital readmission following acute oral corticosteroid (OCS)-treated COPD exacerbations.
We conducted studies from OPCRD (http://optimumpatientcare.org/opcrd) and CPRD (http://www.cprd.com/home/), validated databases for medical research, with linked Hospital Episode Statistics (HES) data for ~20,000 COPD patients aged ≥40 years. For patients with OCS-treated COPD exacerbations treated in primary care, with BEC recorded on first day of OCS treatment (Cohort 1), we assessed treatment failure (COPD-related hospitalisations and OCS prescriptions beyond index OCS course). For patients hospitalised for COPD exacerbations, with BEC measured during an exacerbation-free period during year prior to admission (Cohort 2), we assessed readmission rate. Cox proportional hazards regression analysis adjusted for confounders to estimate association between BEC and treatment outcomes.
Of patients treated with OCS for COPD exacerbations in primary care (Cohort 1), 44% experienced treatment failure following single OCS courses, and 10% (255/2482) were hospitalised ≤6 weeks. Greater BEC was associated with reduced hospital-admission risk (hazard ratio [HR]=0.26 [95% confidence interval [CI]: 0.12, 0.56] per 100-cells/µL increase). BEC increases ≥200 cells/µL from exacerbation-free periods to exacerbations were associated with least hospitalisation risk (HR=0.32 [95% CI: 0.15, 0.71]) vs. no BEC change. For patients hospitalised for COPD exacerbations (Cohort 2), 4-week hospital readmission was 12% (1189/10,245). BEC increases during an exacerbation-free period within the past year were associated with reduced risk of short-term readmission (HR=0.78 [95% CI: 0.63, 0.96]).
Greater BEC predicted better outcomes for patients with OCS-treated COPD exacerbations, whether community or hospital managed. Eosinopenia predicted worse outcomes.