Abstract
BACKGROUND: Biomarkers like sarcopenia, eosinopenia and C-reactive protein (CRP) may predict major adverse events including intubation, ICU admission, mortality and readmission in chronic obstructive pulmonary disease (COPD) exacerbations. We aimed to determine their prognostic utility and accuracy. METHODS: This was a prospective analysis of COPD patients hospitalised for acute exacerbation over one year. Patients with primary diagnoses other than COPD were excluded. Patients were screened to select a sample of 205 participants, with 55 experiencing adverse events including intubation, ICU admission, in-hospital mortality and 30-day readmission. Data on demographics, lung function, symptoms, nutrition, frailty, sarcopenia, eosinophil-to-platelet ratio (EPR) and CRP were extracted. Differences between groups were analysed using t-tests and regression modelling. RESULTS: EPR <0.755 and CRP ≥15.8 mg/dL were significant predictors of adverse events after adjustment, with EPR having an AUC of 0.79 and CRP an AUC of 0.68 for composite outcomes. In multivariate analysis, sarcopenia, EPR and CRP remained significant with the outcome variables (intubation, ICU admission, in-hospital mortality and 30-day readmission). CONCLUSION: EPR and CRP are useful prognostic markers of clinically significant in-hospital outcomes during COPD exacerbations. However, a multidimensional approach may further optimise risk prediction.