Effect of decreasing the frequency of screening for Neisseria gonorrhoeae and Chlamydia trachomatis on the incidence of these infections and antimicrobial use among men who have sex with men using HIV PrEP in Belgium: a retrospective cohort study

Background
There is growing evidence that the benefits of systematic screening for Neisseria gonorrhoeae and Chlamydia trachomatis among men who have sex with men (MSM) using HIV pre-exposure prophylaxis (PrEP) are scarce and that screening leads to high antimicrobial consumption. The PrEP clinic of the Institute of Tropical Medicine (ITM) in Antwerp, Belgium, discontinued systematic screening for these infections in MSM taking PrEP in 2023. In this study, we estimated the effect of reducing the frequency of screening on the incidence of these infections and antimicrobial use among MSM using PrEP.

Methods
We did a retrospective cohort analysis of medical records, laboratory results, and antimicrobial prescriptions (ceftriaxone, doxycycline, and azithromycin) of MSM attending the PrEP clinic of the ITM in Antwerp, Belgium, between Jan 1, 2019, and Dec 31, 2024. We estimated yearly testing rates for N gonorrhoeae and C trachomatis, incidence rates of overall and symptomatic infections, and antimicrobial prescription rates. Results were analysed using Poisson regression, with the number of events as the outcome, years as a continuous predictor and the log(person-time) as offset. Additionally, we did three sensitivity analyses: limiting our study population to individuals engaged in PrEP care in 2019; excluding participants of the Gonoscreen study; and using time as a categorical variable.

Findings
Between Jan 1, 2019, and Dec 31, 2024, 3955 MSM attended the clinic. Over the study period, 11 720 tests were done during 9853·85 person-years. The testing rate decreased significantly from 2267·78 tests per 1000 person-years (95% CI 2176·49–2361·92) in 2019 to 552·81 tests per 1000 person-years (521·53–585·47) in 2024 (yearly rate ratio 0·78, 95% CI 0·77–0·78; p<0·0001). The incidence of N gonorrhoeae decreased from 170·84 cases per 1000 person-years (95% CI 146·47–198·11) in 2019 to 85·81 cases per 1000 person-years (73·77–99·27) in 2024 (yearly incidence rate ratio [IRR] 0·91, 95% CI 0·88–0·94; p<0·0001). In the same period, the incidence of C trachomatis decreased from 198·17 cases per 1000 person-years (95% CI 171·85–227·39) in 2019 to 60·69 cases per 1000 person-years (50·63–72·16) in 2024 (yearly IRR 0·84, 95% CI 0·81–0·87; p<0·0001). There was no increase in the incidence of symptomatic N gonorrhoeae (IRR 0·98, 95% CI 0·92–1·03), C trachomatis (0·95, 0·88–1·01), or lymphogranuloma venereum (0·95, 0·84–1·09). Ceftriaxone, doxycycline, and azithromycin prescriptions decreased (rate ratio 0·95, 95% CI 0·91–0·99; p=0·023, 0·47, 0·43–0·52; p<0·0001, and 0·90, 0·87–0·94; p<0·0001, respectively). Similar results were found in the first sensitivity analysis. In the sensitivity analysis excluding participants of the Gonoscreen study, rates of ceftriaxone and doxycycline prescriptions decreased between 2019 and 2024 but became non-significant. In the third sensitivity analysis, using time as a categorical variable, the prescription rates of ceftriaxone were lower in 2024 compared with 2019 and the incidence rates of symptomatic C trachomatis were significantly lower in 2024 compared with 2019.

Interpretation
The incidence of symptomatic N gonorrhoeae and C trachomatis infections did not increase over time despite less frequent screening and antimicrobial prescriptions decreased. Our findings contribute to the evidence supporting the potential of reducing the frequency of screening as an antimicrobial stewardship intervention among MSM using PrEP. Future studies are needed to confirm our findings.

Funding
None.