Data from 8 individual RCTs (3,715 patients) were included. Three of these studies were reported in one secondary analysis paper.
ICS significantly reduced the rate of decline in FEV1 by 7.7 mL/year (95% CI: 1.3, 14.2, P=0.02), based on 5 plus 3 combined studies. No significant heterogeneity was detected (P=0.32).
There was a greater reduction in the rate of decline of FEV1 in studies using high-dose ICS (4 RCTs, 2,416 patients); the reduction was 9.9 mL/year (95% CI: 2.3, 17.5, P=0.01). No significant heterogeneity was detected (P=0.50).
There was a non significant trend towards a reduction in the rate of decline in patients with a baseline FEV1 of 50% or less of that predicted (1,032 patients); the reduction was 18.3 mL/year (95% CI: -1.5, 38.0, P=0.07), based on 2 plus 3 combined studies. No significant heterogeneity was detected (P=0.25).
The asymmetrical funnel plot and Eggar test (P=0.03) suggested the possibility of publication bias.