Objective A multicentre, randomised non-inferiority trial compared the efficacy and safety of 14 days of ofloxacin and metronidazole (standard-of-care (SoC)) versus a single dose of intramuscular ceftriaxone followed by 5 days of azithromycin and metronidazole (intervention arm (IA)) in women with mild-to-moderate pelvic inflammatory disease (PID).

Methods Women with a clinical diagnosis of PID presenting at sexual health services were randomised to the SoC or IA arms. Treating clinicians and participants were not blinded to treatment allocation but the clinician performing the assessment of primary outcome was blinded. The primary outcome was clinical cure defined as ≥70% reduction in the modified McCormack pain score at day 14–21 after starting treatment. Secondary outcomes included adherence, tolerability and microbiological cure.

Results Of the randomised population 72/153 (47.1%) reached the primary end point in the SoC arm, compared with 68/160 (42.5%) in the IA (difference in cure 4.6% (95% CI −15.6% to 6.5%). Following exclusion of 86 women who were lost to follow-up, attended outside the day 14–21 follow-up period, or withdrew consent, 72/107 (67.3%) had clinical cure in the SoC arm compared with 68/120 (56.7%) in the IA, giving a difference in cure rate of 10.6% (95% CI −23.2% to 1.9%). We were unable to demonstrate non-inferiority of the IA compared with SoC arm. Women in the IA took more treatment doses compared with the SoC group (113/124 (91%) vs 75/117 (64%), p=0.0001), but were more likely to experience diarrhoea (61% vs 24%, p<0.0001). Of 288 samples available for analysis, Mycoplasma genitalium was identified in 10% (28/288), 58% (11/19) of which had baseline antimicrobial resistance-associated mutations.

Conclusion A short-course azithromycin-based regimen is likely to be less effective than the standard treatment with ofloxacin plus metronidazole. The high rate of baseline antimicrobial resistance supports resistance testing in those with M. genitalium infection to guide appropriate therapy.

Trial registration number 2010-023254-36.