Introduction There is increasing concern about treatment failure among those treated for anogenital chlamydia infection. We used genotyping and survey data to differentiate between reinfection  and treatment failure among men who have sex with men (MSM), heterosexual men and women, diagnosed with repeat chlamydia infection within 1- 4 months after treatment with azithromycin.

Methods Participants completed an online survey capturing treatment and sexual behaviour data since initial diagnosis. Specimens from initial and repeat infections were included in the study. Chlamydia serovars were determined using quantitative PCR assays. When the same serovar was detected in both specimens for participants, MLST was used to further discriminate between genotypes. An algorithm based on genotype and sexual behaviour data was used to differentiate treatment failure from reinfection.

Results There were 600 participants (200 MSM, 200 heterosexual males, 200 females) diagnosed with chlamydia. Of 301/600 who retested between 1–4 months: 258/301 (85.7%) were cured (treated and negative on retest); 4/301 (1.3%) had a definite reinfection (positive retest and different genotype); 19/301 (6.3%) had probable reinfection (positive retest, same genotype and reported unprotected sex with the same or a different partner); 17/301 (5.6%) had possible treatment failure (positive retest, same genotype and reported not having sex or always using condoms); 1/301 (0.3%) had a persistent infection (positive retest, same genotype and no documented treatment); and 2/301 (0.7%) could not be categorised due to insufficient information. Possible treatment failures were more common in MSM (11.3%, 12/106) vs other groups (2.6%, 5/195; p < 0.01). Among the possible treatment failures in MSM, 10/12 (83.3%) were initial rectal samples.

Conclusion Treatment failure was common in MSM with rectal chlamydia, suggesting the need for treatment efficacy trials.

Disclosure of interest statement No conflict of interest is declared.