Without protective immunity, recurrent sexually transmitted infections (STI) could occur. In this study, we retrospectively collected STI diagnosis records from public STI clinics attended by an average of 6,000 male patients annually in Hong Kong in 2009–2019. We estimated the prevalence of three bacterial STI (syphilis, chlamydia and gonorrhoea) coinfection from 2009 to 2019, and examined the factors associated with coinfection in 2014/15 and repeat infection in 2009–2019. We observed an increasing coinfection prevalence in male attendees with bacterial STI over the years, which reached the highest level of 15% in 2019. Among 3,698 male patients in 2014–2015, chlamydia/gonorrhoea coinfection was the commonest among all coinfections (77%). Factors such as young age (29 or below), HIV-positive status, and a history of concurrent genital warts/herpes were positively associated with coinfection in 2014/15 in multivariable logistic regression. Of all male patients with STI coinfection in 2014/15, those of age 30–49 and who self-reported as men who have sex with men (MSM) were more likely to have been repeatedly infected in 2009–2019. The results support the implementation of regular multi-STI testing as an STI control strategy for selected communities like MSM and people living with HIV.

Repeat rectal chlamydia infection is common in men who have sex with men (MSM) following treatment with 1 g azithromycin. This study describes the association between organism load and repeat rectal chlamydia infection, genovar distribution, and efficacy of azithromycin in asymptomatic MSM. Stored rectal chlamydia-positive samples from MSM were analysed for organism load and genotyped to assist differentiation between reinfection and treatment failure. Included men had follow-up tests within 100 days of index infection. Lymphogranuloma venereum and proctitis diagnosed symptomatically were excluded. Factors associated with repeat infection, treatment failure and reinfection were investigated. In total, 227 MSM were included – 64 with repeat infections [28·2%, 95% confidence interval (CI) 22·4–34·5]. Repeat positivity was associated with increased pre-treatment organism load [odds ratio (OR) 1·7, 95% CI 1·4–2·2]. Of 64 repeat infections, 29 (12·8%, 95% CI 8·7–17·8) were treatment failures and 35 (15·4%, 95% CI 11·0–20·8) were reinfections, 11 (17·2%, 95% CI 8·9–28·7) of which were definite reinfections. Treatment failure and reinfection were both associated with increased load (OR 2·0, 95% CI 1·4–2·7 and 1·6, 95% CI 1·2–2·2, respectively). The most prevalent genovars were G, D and J. Treatment efficacy for 1 g azithromycin was 83·6% (95% CI 77·2–88·8). Repeat positivity was associated with high pre-treatment organism load. Randomized controlled trials are urgently needed to evaluate azithromycin's efficacy and whether extended doses can overcome rectal infections with high organism load.