Twelve RCTs (n=4,576) were included in the review.
All of the included studies were of a high quality (Jadad score of at least 4).
Primary outcomes. There was a significant reduction in the number of patients withdrawing due to asthma in the salmeterol-ICS group compared with the high-dose ICS group: 59 out of 2,036 versus 86 out of 1,992. The OR was 1.58 (95% CI: 1.12, 2.24; 10 studies).
There was a significant reduction in the number of patients with one or more moderate or severe exacerbations in the salmeterol-ICS group compared with the high-dose ICS group: 184 out of 2,312 versus 243 out of 2,264. The OR was 1.35 (95% CI: 1.10, 1.66; 12 studies).
Secondary outcomes. Significant heterogeneity was found for morning PEF, FEV1, daytime beta-agonist use and night-time beta-agonist use. Morning and evening PEF were significantly greater and daytime use of beta-agonist was significantly less for patients taking salmeterol-ICS than for those taking high-dose ICS.
FEV1 (using a random-effects model) was barely statistically significantly increased for patients taking salmeterol-ICS compared with high-dose ICS. There was no significant difference between treatments in night-time awakenings and night-time beta-agonist use.