Eight RCTs (n=587) were included, of which seven were published and one was a conference abstract.
Five RCTs scored 4 out of a possible 5 points for quality on the modified Jadad scale and three scored 2 points. All RCTs were described as randomised and provided details of withdrawals. Five reported the generation of random numbers and concealment of allocation. Two RCTs were single-blinded; none were double-blinded. No publication bias was detected in any analysis.
There was no statistically significant difference in treatment success between the groups when azithromycin was compared with erythromycin (OR 2.66, 95% CI: 0.69,10.29; 4 RCTs, n=293) or erythromycin/amoxicillin (OR 1.45, 95% CI: 0.82, 2.57; 6 RCTs, n=344). Cure rates ranged from 63 to 100% for azithromycin, from 72 to 94% for erythromycin, and from 58 to 80% for amoxicillin.
There were fewer GI adverse events in the azithromycin group than in the erythromycin group (OR 0.11, 95% CI: 0.07, 0.18; 6 RCTs, n=374) or the erythromycin/amoxicillin group (OR 0.16, 95% CI: 0.06, 0.4; 7 RCTs, n=412). There was no statistically significant difference between the groups in the incidence of other adverse events when azithromycin was compared with erythromycin (5 RCTs, n=334) or with erythromycin/amoxicillin (7 RCTs, n=482). There were significantly fewer total adverse events in the azithromycin group than either the erythromycin group (OR 0.11, 95% CI: 0.07, 0.18; 5 RCTs, n=289) or the erythromycin/amoxicillin group (OR 0.13, 95% CI: 0.08, 0.21; 6 RCTs, n=325).
There were significantly fewer withdrawals in the azithromycin group than in the erythromycin group (OR 0.12, 95% CI: 0.04, 0.37; 4 RCTs, n=203). In three of these RCTs, withdrawals comprised women who were unable to tolerate the intervention because of severe adverse effects, while in the fourth the dose of erythromycin was reduced to 250 mg to allow women to complete the protocol. There was no statistically significant difference in withdrawals between the groups when azithromycin was compared with erythromycin/amoxicillin (6 RCTs, n=352).
Women in the azithromycin group were more compliant with therapy than those in the erythromycin group (OR 23.7, 95% CI: 9.34, 60.14; 6 RCTs, n=374) or in the erythromycin/amoxicillin group (OR 21.96, 95% CI: 9.05, 53.3; 7 RCTs, n=413).
There was no significant difference in neonatal outcomes between the groups when azithromycin was compared with erythromycin (2 RCTs, n=246).