Risk factors for corticosteroid- and antibiotic only-treated asthma attacks in the NOVELTY cohort

Abstract

<b>Background:</b> Risk factors for asthma attacks requiring corticosteroid treatment include elevated biomarkers of type-2 airway inflammation. In real life, asthma attacks are often treated with antibiotics, and little is known about these attacks. <b>Aim:</b> We explored predictors for corticosteroid- and antibiotic only-treated attacks in the multi-country, prospective, observational NOVELTY cohort (NCT02760329). <b>Methods:</b> Patients with physician-assigned asthma with baseline data for 15 candidate predictors (including blood eosinophils [EOS] and fractional exhaled nitric oxide [FeNO]) and data for exacerbation history (acute asthma requiring ≥3 days of corticosteroids and/or hospitalisation, or antibiotics only) in the 12 months prior to and the 12 months post-baseline, not on biologics, were included. Adjusted rate ratios [95% confidence intervals] were calculated to determine risk factors for annualised corticosteroid- and antibiotic only-treated attacks. <b>Results:</b> Of 4,753 patients with asthma, 961 with full predictors and outcomes data were included. Significant predictors for corticosteroid-treated attacks were female sex (1.54 [1.08–2.21]), increased symptoms (Asthma Control Test 0.94 [0.91–0.97], for one unit) and a prior corticosteroid-treated attack (3.68 [2.69–5.03]); but not EOS and FeNO. Predictors for antibiotic only-treated attacks were low FEV₁% (0.98 [0.96–1.00], for one unit), comorbid rhinosinusitis (2.42 [0.98–5.93]) and a prior antibiotic only-treated attack (4.24 [1.53–12.07]). <b>Conclusion:</b> Risk factors for corticosteroid- and antibiotic only-treated attacks differed. Contrary to clinical trial reports, type-2 biomarkers did not predict asthma attacks in this subset of patients.