Smoking influences response to inhaled corticosteroids in patients with asthma: a meta-analysis.

Objective: To determine the effects of smoking on efficacy of inhaled corticosteroids in asthmatics. Research design and methods: Prospective/retrospective, controlled studies published up to May 2012 in smokers versus non-smokers with asthma using inhaled corticosteroids (ICSs) were retrieved from PubMed, Highwire and Embase databases. Main outcome measures: The primary outcomes were pulmonary function parameters [FEV1 (forced expiratory volume in 1 second), a.m. PEF (a.m. peak expiratory flow) and p. m. PEF (p.m. peak expiratory flow)]. Secondary outcomes included differential eosinophil counts in sputum and peripheral blood, asthma control questionnaire-5 (ACQ-5) and concomitant use of medication. Pooled SMDs (standardized mean differences) were estimated by using a random-effect model and a fixed-effect model based on between-study heterogeneity. Results: Of 242 studies identified, ten (N = 4070) met the inclusion criteria (seven of budesonide-based interventions, two of fluticasone and one unknown ICS trial). Both smokers and non-smokers were similar in age and baseline FEV1. Smoking yielded considerably reduced mean change in FEV1 [SMD = -0.197, 95% CI: (-0.327, -0.066), p = 0.003], morning PEF [SMD = -0.796, 95% CI: (-1.047, -0.545), p<0.001], night-time PEF [SMD = -0.501, 95% CI: (-0.797, -0.204), p = 0.001] and post-treatment FEV1 [SMD = -0.178, 95% CI: (-0.309, -0.046), p = 0.008] and increased use of concomitant medications [SMD = -0.537, 95% CI: (0.166, 0.908), p = 0.005] in smokers, but not non-smokers with asthma, although there was no statistical difference in allergy-related endpoints and asthma score (ACQ-5). Conclusion: Smoking is associated with attenuated response to inhaled corticosteroids in asthmatics. This offers a rationale for persuasion of smoking cessation. Treatment outcomes might be improved by smoking cessation leading to improved lung function and ameliorated symptoms.