Relative efficacy and safety of inhaled corticosteroids in patients with asthma Systematic review and network meta-analysis

Background: Inhaled corticosteroids (ICSs) are recommended as first-line controller medications for persistent asthma. However, guidelines on the initial ICS doses, step-up and step-down algorithms, and when to switch to combination therapy vary. Objective: To understand the ideal starting doses of ICS therapy based on current evidence and to systematically compare low, moderate, and high starting doses of ICSs as monotherapy and in combination with long-acting beta-agonists with respect to efficacy and safety. Methods: MEDLINE, Embase, and Cochrane databases were searched for relevant English-language articles published from 1980 to November 17, 2018. Randomized controlled trials with adult, steroid-naive, ICS-free (for >= 4 weeks) patients with asthma and a duration of 4 weeks or longer with an ICS treatment arm (monotherapy or combination therapy) were included. Separate fixed-effects Bayesian network meta-analyses were conducted on the extracted data for peak expiratory flow, forced expiratory volume in 1 second, nighttime rescue medication use, nighttime symptom score, and study withdrawal because of an adverse event. Results: A total of 31 randomized controlled trials were analyzed. All starting doses of ICSs were comparable with respect to nighttime rescue medication use, nighttime symptom score, change in forced expiratory volume in 1 second, and study withdrawal because of an adverse event. Significant improvement in morning peak expiratory flow was observed with high-dose ICSs and with low- and moderate-dose ICSs and long-acting beta-agonists than with low-dose ICSs. Conclusion: Overall, a high starting dose of ICSs had no additional clinical benefit in 3 of the 4 efficacy parameters compared with low or moderate ICS doses for controlling moderate to severe asthma but might have potential safety concerns. (C) 2020 American College of Allergy, Asthma & Immunology. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).