Inhaled corticosteroids and incident pneumonia in patients with asthma: Systematic review and meta-analysis

To systematically review all available studies on inhaled corticosteroid use and incident pneumonia in asthma patients.We performed a literature search from January 1, 1993, through August 15, 2015, using PubMed, Medline, CENTRAL, EMBASE, Scopus, ISI, Regulatory Documents, Web of Science and manufacturers' web clinical trial registries with multiple search terms. We included studies that compared the risk of incident pneumonia among patients utilizing and not utilizing inhaled corticosteroids. We then summarized risk estimates into two random-effect meta-analyses; one including randomized controlled trials and another one including observational studies.Fourteen studies were estimable; ten randomized controlled trials included 19,098 participants and four observational studies included 44,016 participants. There was no heterogeneity in randomized trials and summed risk ratio demonstrated the use of inhaled corticosteroids was protective of pneumonia; risk ratio 0.74, 95% CI 0.57 to 0.95, p=0.02. On the contrary, observational studies showed summed odds ratio of 1.97; 95% CI 1.87 to 2.07, p<0.0001, I²=0%, suggesting increased risk of pneumonia with use of inhaled corticosteroids in asthma patients.Inhaled corticosteroids are associated with decreased risk of incident pneumonia in patients with asthma based on meta-analysis of available randomized trials. Although observational studies in similar patients suggested higher risk of pneumonia, the inherent methodological limitations confer lower grade of confidence in these studies.