Abstract
BACKGROUND: The association between blood eosinophils and COPD exacerbation has been controversial. This study aims to investigate whether high blood eosinophils predict the risk of COPD exacerbation across different thresholds and subgroups. METHODS: PubMed, Embase and Web of science were searched for randomized controlled trial (RCT) and observational studies regarding the relationship between blood eosinophils and COPD exacerbation. Pooled risk ratio (RR) for COPD exacerbation was calculated using the Mantel-Haenszel method with a random-effects model. RESULTS: A total of 21 studies (1 RCT and 20 observational studies) with 79868 participants were included. Thresholds of high blood eosinophils including absolute counts (200, 300 and 400 cell/μL) and percentages (2%, 3% and 4%) were analyzed respectively. Pooled analyses suggested that high blood eosinophils were significantly associated with increased risk of COPD exacerbation when using the thresholds of 300 cells/μL (RR 1.21, 95%CI 1.12-1.30, P <0.001, 16 studies), 400 cells/μL (RR 1.79, 95%CI 1.41-2.28, P <0.001, 3 studies), 2% (RR 1.26, 95%CI 1.02-1.55, P = 0.030, 10 studies) and 4% (RR 1.44, 95%CI 1.05-1.96, P = 0.022, 4 studies), but not 200 cells/μL and 3% (P>0.05). Moreover, high blood eosinophils contributed to moderate-severe exacerbation of COPD by the cutoffs of 300 cells/μL (RR 1.30, 95%CI 1.16-1.45, P<0.001, 11 studies) and 2% (RR 1.33, 95%CI 1.02-1.76, P = 0.037, 8 studies). In subgroup analyses, the pooled results further showed a significant association between high blood eosinophils (especially over 300 cells/μL) and risk of COPD exacerbation among patients from Europe and Asia, and whether with stable or exacerbation phase at baseline, and regardless of the follow-up time (≤ or > 1year). CONCLUSIONS: This study demonstrates that high blood eosinophils (over 300 cells/μL or 2%) could predict the risk of moderate-severe exacerbation of COPD in specific subgroups. However, large sample-sized, prospective, and well-designed studies are required to validate the present findings.