Four RCTs were included in the review (n=368). All of the trials had a Jadad score of 4 (no further details were reported).
Azithromycin was associated with a significant increase in FEV1 (MD 3.53%, 95% CI 0.00 to 7.07; four RCTs) and FVC compared with placebo (MD 4.24%, 95% CI 2.02 to 6.45; four RCTs). There was some heterogeneity for FEV1 (I2 38%) and none for FVC. For FVC the difference between groups was not significant in trials with a low proportion of Pseudomonas.
No significant FEV1 or FVC changes were found at one to three months follow-up.
Large amounts of heterogeneity meant studies could not be combined for secondary efficacy outcomes. Three trials reported a significant decrease in the number of antibiotic courses (no further details were reported).
The risk of gastrointestinal effects was higher with azithromycin (RR 1.72 95% CI 1.33 to 2.21). The main side effects reported were nausea (RR 2.04, 95% CI 1.19 to 3.45) and diarrhoea (RR 2.12, 95% CI 1.10 to 4.08), which were significantly higher with azithromycin than with placebo.
There was no evidence of publication bias (p=0.31).