Anorectal Neisseria gonorrhoeae infections in women with and without reported anal sex and sex workers in sexual health centres in the Netherlands: a retrospective cohort study

Background Control of Neisseria gonorrhoeae infection (gonorrhoea) depends on effective testing strategies. Anorectal testing in women is often done on indication of anal sex; however, anorectal infections are seen with and without anal exposure, possibly caused by autoinoculation. This study aims to enhance understanding of anorectal infections in women, by identifying risk factors for anorectal diagnosis. Methods In this retrospective cohort study we used national surveillance data from Dutch sexual health centres from Jan 1, 2016, to Dec 31, 2021. We included cisgender women having sex with men who were tested urogenitally and anorectally for gonorrhoea. Due to different testing policies, we identi fi ed three groups: women who had not reported recent anal sex (in the past 6 months), women who had reported recent anal sex, and sex workers. Extracted data for analyses included demographics, sexual behaviour, and diagnosis of a sexually transmitted infection (STI). Per group, multivariable models using Firth ' s penalised maximum likelihood logistic regression were constructed, identifying determinants of anorectal gonorrhoea among all women and among gonorrhoea-positive women only. Variables included in model construction were age, education level, migration background, number of partners, condom use, partner notification, STI symptoms, having a partner who has sex with men (MSM) or a migrant partner, previous STI test, anal sex, and chlamydia and gonorrhoea diagnoses per anatomical location. Findings In total, 117 693 women were included: 43 757 women without reported recent anal sex, 51728 women with reported recent anal sex, and 22 208 sex workers. In all three groups, around 2% of women were gonorrhoea positive, and 70% or more of women had an anorectal infection. The strongest determinant of anorectal gonorrhoea was a concurrent urogenital gonorrhoea diagnosis (adjusted odds ratios [aOR] 782 [95% CI 605 - 1018]) among women without reported recent anal sex (612 [490 - 768] among women with reported recent anal sex, and 464 [335 - 652] among sex workers). Among gonorrhoea-positive women, determinants of anorectal gonorrhoea were urogenital and anorectal chlamydia co-infection (aOR 2 . 03 [95% CI 1 . 38 - 3 . 02], for women without reported anal sex) and migration background (1 . 44 [1 . 02 - 2 . 06], for women with reported anal sex). Determinants among sex workers were condomless sex (2 . 43 [1 . 55 - 3 . 82]), anal sex (1 . 71 [1 . 10 - 2 . 66]), MSM or migrant partner (1 . 78 [1 . 13 - 2 . 79]), and urogenital and anorectal chlamydia co-infection (2 . 28 [1 . 11 - 5 . 14]). Interpretation These fi ndings support the possibility of an autoinoculation process from the urogenital to the anorectal location due to the very strong correlation between urogenital and anorectal gonorrhoea, and due to the similarity of results across all three groups. Current testing strategies could miss anorectal infections, which should be considered when developing gonorrhoea prevention and control guidelines. Funding None.