Medline, PubMed, Embase and the Cochrane Controlled Trials Register were searched till end 2012. Inclusion criteria included 1) randomised controlled trial of azithromycin versus doxycycline for the treatment of urethral or cervical chlamydia, and; 2) evaluation of microbial cure within 3 months of treatment. Type of diagnostic test, duration of follow up, gender, patient status (symptomatic versus other) and microbial cure were extracted. The primary outcome was efficacy difference (ED=doxycycline efficacy minus azithromycin efficacy) at final follow up. Meta-analysis calculated a pooled efficacy for each treatment and the difference in efficacy between treatments.
Of 692 references identified, 23 trials met the inclusion criteria. 1099 individuals were treated with azithromycin and 876 with doxycycline; all studies reported efficacy within 6 weeks follow-up. Pooled cure rates were 94.4% (95%CI:91.9%, 96.9%) for azithromycin and 97.5% (95%CI:96.3%,98.8%) for doxycycline. The pooled efficacy difference was 2.7% (95%CI:0.7%,4.8%) showing a significant difference in favour of doxycycline; there was negligible heterogeneity between studies (I2 = 0.0%, p=0.46). There was no significant efficacy difference in men (ED=4.3%;95%CI:- -0.2%, 8.9%) or women (ED=1.0%;95%CI: -4.4%, 6.5%). When stratified by type (ED=4.3%;95%CI:- -0.2%, 8.9%) or women (ED=1.0%;95%CI: -4.4%, 6.5%). When stratified by type of test, efficacy was significantly higher for doxycycline in NAAT-based studies (ED=5.4%;95%CI: 0.6%, 10.2%), but not in culture-based studies (ED=1.9%;95%CI: -0.4%, 4.1%). Efficacy was higher for doxycycline in symptomatic individuals (ED=5.8%;95%CI: 1.0%, 10.5.%)
Conclusion: These results suggest that doxycycline may be more effective than azithromycin for the treatment of urethral or cervical chlamydia infection, especially for symptomatic individuals. Efficacy however needs to be balanced against compliance, with a 7-day course of doxycycline.