Biomarker defined infective and inflammatory asthma exacerbation phenotypes in hospitalized adults: clinical impact and phenotype stability at recurrent exacerbation

Objective: Acute exacerbations (AEs) of asthma are heterogeneous in terms of triggers, outcomes, and treatment response. This study investigated biomarker defined infective and inflammatory AE phenotypes in hospitalized adult asthma patients, and their impact on clinical outcomes and phenotype stability at AE recurrence. Method: Patients with asthma admitted with an AE between January 2010 and December 2011 with a 3-year follow-up were retrospectively studied. AEs were categorized into infective (CRP >10 mg/L) vs non-infective, eosinophilic (blood eosinophils >= 0.2 x 10(9) cells/L) vs non-eosinophilic, and viral (CRP >10 to <40 mg/L) vs bacterial (CRP >= 40 mg/L) phenotypes. Clinical impact of the index AE, the risk and time to a second AE and AE phenotype stability were analyzed using Kaplan-Meier survival curves and McNamar's test. Result: 294 asthma patients were included: 47% had infective AE with a longer length of stay than non-infective AE (2.0 vs. 1.0 days, p = 0.01). The proportion of patients with eosinophilic AEs was evenly distributed across infective and non-infective AE (40% vs. 46%), although more patients with viral had eosinophilia than bacterial AE (46% vs. 26%). During follow-up, 18% had recurrent AE; with a higher risk in viral AE than bacterial AE (25% vs. 8%, p = 0.02). Both inflammatory and infective AE phenotype were stable at recurrent AE. Conclusion: AE phenotyping in hospitalized asthma patients, based on CRP and blood eosinophils, revealed prolonged hospital stay in infective AEs and a higher risk of recurrent AE requiring hospitalization in viral versus bacterial AEs. Moreover, infective, and inflammatory AE phenotypes were rather stable at recurrent AE. Our results suggest a role for biomarker guided phenotyping of AEs of asthma.