Introduction

Neisseria gonorrhoeae (Ng) is one of the most common bacterial sexually transmitted infections (STIs).1 In the Netherlands, the number of Ng cases has been increasing since 2022, with an increase of 38% in women and 20% in heterosexual men in the first half of 2023.2–4 The recent increase in Ng among women and heterosexual men was also observed in other European countries.5 This is noteworthy given that Ng previously had a low prevalence in these populations, and prevention efforts were primarily aimed at gay, bisexual and other men who have sex with men (GBMSM).2

During the COVID-19 pandemic, STI screening was substantially reduced at all Centres for Sexual Health (CSH) of the Public Health Services of the Netherlands; only individuals with symptoms, a partner notification for an STI, an indication for HIV post-exposure prophylaxis or a history of sexual violence were eligible for testing. This led to artificial increases in positivity rates. While it was expected that positivity rates would decline as STI screening expanded again, recent data show the opposite.3 6 Reduced condom use, increase in sexual partners, frequent partner changes and the removal of COVID-19 restrictions have been found associated with an increased risk of Ng.5 Prior to the COVID-19 era, Ng was primarily concentrated in cisgender women and heterosexual men who were vocationally educated and of non-Dutch origin. However, highly educated individuals under the age of 25 years of Dutch origin have also become affected after COVID-19 measures ended.4 Given the rapid increase of Ng in populations with previously low Ng positivity rates, examining determinants associated with this increase is important.

The aim of this study was to assess the positivity rate of Ng and symptoms suggestive for Ng pre-COVID-19, during and post-COVID-19 pandemic among: (1) men who exclusively have sex with men (MSM), (2) men who exclusively have sex with women (MSW), (3) men who have sex with men and women (MSMW), (4) women and (5) transgender and gender diverse people who attended the CSH in Amsterdam. To examine determinants of any changes in Ng positivity rate, we also analysed whether changes were only observed in people with specific behavioural factors. As a rise in Ng may also lead to a rise in complications of Ng, likepelvic inflammatory disease (PID) or epididymitis, we also assessed the temporal changes in the diagnoses of these two conditions.

MethodsStudy design and participants

We used routinely collected data from electronic patient records of all individuals tested for Ng at the CSH of Amsterdam. GBMSM, transgender and gender diverse people, and sex workers are screened for STI regardless of age. Cisgender women and heterosexual cisgender men <25 years can also test regardless of symptoms or partner notification. Cisgender women and heterosexual cisgender men ≥25 years are only screened for STI if they have symptoms or a partner notification.

We assessed the positivity rate of Ng, symptoms indicative for Ng and the incidence of PID and epididymitis in three time periods: pre-COVID-19 (1 January 2018–12 March 2020), during COVID-19 (13 March 2020–1 March 2022) and post-COVID-19 (2 March 2022–31 December 2023). We distinguished as in 2020, and to a lesser extent in 2021, sexual healthcare, including STI screening, was substantially downscaled and STI screening only returned to its normal capacity after March 2022.

For this study, we included consultations between 1 January 2018 and 31 December 2023. Individuals could contribute multiple data points if they had more than one consultation during the study period.

Procedures

The anatomical location(s) sampled for each group is summarised in table 1. Symptomatic individuals were treated for presumptive Ng during the initial consultation based on a positive Gram-stained smear. Ng-related symptoms were urethral, vaginal anal and/or conjunctival discharge, burning sensation in anogenital area and/or during miction, pain in testicles, sore throat, genital ulcers, vaginal contact bleeding and/or interim blood loss and abdominal pain. Women with abdominal pain in combination with cervical motion tenderness, painful or swollen adnexa, or fever were suspected of having PID, whereas men with (unilateral) scrotal pain and swelling were suspected of having epididymitis. Asymptomatic individuals were screened for Ng during the initial consultation using Nucleic Acid Amplification Test and, if positive, treated during a follow-up consultation. Treatment was provided free-of-charge and involved 1 g of intramuscular ceftriaxone monotherapy. Routinely collected demographic (eg, age and country/region of birth) and clinical data (eg, HIV status, antiretroviral therapy and recent antibiotic use), as well as information on sexual behaviour (eg, number of sexual partners, consistency of condom use and engagement in chemsex), were extracted from the electronical medical records.

Table 1

Anatomical location tested for Ng per study group

Statistical analysis

We calculated the Ng positivity rate per calendar year and per COVID-19 period for any Ng (pharyngeal, anal and/or urogenital Ng) by dividing the total number of Ng diagnoses by the sum of all consultations in the period. We also assessed the annual Ng positivity rate for each anatomical location. All analyses were stratified by a combination of gender and sexual preference (ie, MSM, MSW, MSMW, women and transgender and gender diverse people).

We modelled the Ng positivity rate over time using relative risk regression (ie, generalised linear model with a Poisson distribution, log link and robust variance estimations). We compared the relative difference in Ng positivity rates between periods by including a covariate for period (ie, pre-COVID-19, during, and post-COVID-19 pandemic periods). Since multiple observations could occur within individuals, we included a random intercept for the individual to account for within-individual variability. If the Ng positivity rate was significantly higher in the post-COVID-19 compared with pre-COVID-19 period, we further stratified analysis on various factors by including covariates for the factor and the interaction between time period and factor. The following factors were examined: number of sexual partners in the last 6 months (categorised as 0–1, 2–3, 4–6, ≥7); consistency of condom use during vaginal and/or anal sex in the last 6 months (ie, condoms were used during all reported sex acts in the last 6 months vs condoms were used during some or none of the sex acts); engagement in chemsex (ie, sex under the influence of crystal methamphetamine, mephedrone or gamma hydroxybutyrate/gamma butyrolactone or ketamine) in the last 6 months; sex under the influence of other drugs (excluding alcohol and chems (drugs used in chemsex)) in the last 6 months; self-reported history of a bacterial STI in the last 12 months. We assessed changes over time in individuals with Ng reporting symptoms in a similar manner.

We modelled the number of diagnosis over time of (1) PID among women and transgender and gender diverse people with a cervix and (2) of epididymitis among MSM, MSW, MSMW and transgender and gender diverse people with a penis using a similar approach. Given the low incidence of PID and epididymitis at the CSH, and the frequent absence of a causative pathogen, we examined all cases of PID and epididymitis, regardless of association with Ng.

We considered a p value of <0.05 as significant. All analyses were performed using Stata (V.17.0; College Station, Texas).

Results

Between 1 January 2018 and 31 December 2023, there were 304 077 consultations among 112 802 individuals during which Ng testing was performed. Of these consultations, 129 500 were done with MSM, 45 972 with MSW, 10 951 with MSMW, 113 820 with women and 3834 with transgender and gender diverse people. The median number of consultations was 2 (IQR=1–5). Of the 304 077 consultations, 110 878 (36.5%) were done pre-COVID-19, 84 917 (27.9%) during COVID-19 and 108 282 (35.6%) post-COVID-19.

At first consultation captured in the study period, MSW and women were younger and had fewer sexual partners than MSM, MSMW and transgender and gender diverse people (table 2). The majority of MSM, MSW, MSMW and women were Dutch and had completed tertiary education, whereas the majority of transgender and gender diverse people were born in Latin America and had completed none, primary or another form of education.

Table 2

Sociodemographic and behavioural characteristics at first consultation captured within the study period of 25 032 MSM, 27 612 MSW, 4637 MSMW, 54 482 women and 1039 transgender and gender diverse persons who attended STI screening at the Center for Sexual Health, Amsterdam, the Netherlands, 1 January 2018 to 31 December 2023

Changes in Ng post-COVID-19

The Ng positivity rate increased for MSW, MSMW and women post-COVID-19 compared with pre-COVID-19 (figure 1, table 3). Among MSM, the Ng positivity rate was 12.4% (95% CI 12.1% to 12.7%) pre-COVID-19 and 13.0% (95% CI 12.7% to 13.3%) post-COVID-19, whereas for transgender and gender diverse people, the Ng positivity rate was 8.1% (95% CI 6.3% to 10.3%) pre-COVID-19 and 10.4% (95% CI 9.1% to 11.9%) post-COVID-19. The Ng positivity rate increased from 1.7% (95% CI 1.5% to 1.9%) to 2.3% (95% CI 2.1% to 2.5%) in MSW, and from 6.7% (95% CI 5.9% to 7.5%) to 9.9% (95% CI 9.0% to 10.8%) in MSMW. In women, the Ng positivity rate almost doubled, from 1.3% (95% CI 1.2% to 1.4%) pre-COVID-19 to 2.9% (95% CI 2.8% to 3.1%) post-COVID-19.

Figure 1

Positivity rate of Neisseria gonorrhoeae over time among MSM, MSW, MSMW, women and transgender and gender diverse persons, Center for Sexual Health, Amsterdam, the Netherlands, 1 January 2018 to 31 December 2023. Abbreviations: MSM, men who have exclusively sex with men; MSW, men who have exclusively sex with women; MSMW, men who have sex with men and women. Note: Bars represent the number of consultations, while lines depict Ng positivity rates.

Table 3

Changes in Ng positivity pre-COVID-19, during and post-COVID-19, Center for Sexual Health, Amsterdam, the Netherlands, 1 January 2018 to 31 December 2023

For MSW, the positivity rate of urogenital Ng increased from 1.7% pre-COVID-19 to 2.3% post-COVID-19 (online supplemental figure 1). For MSMW, the positivity rate of pharyngeal, anal and urogenital Ng increased from 3.5%, 4.2% and 2.1% pre-COVID-19 to 6.3%, 6.5% and 2.8% post-COVID-19, respectively. For women, the positivity rate for pharyngeal, anal and urogenital Ng increased from 1.5%, 1.2% and 0.9% pre-COVID-19 to 3.5%, 2.4% and 2.2% post-COVID-19, respectively.

The relative risk for Ng increased post-COVID-19 for MSW (RR=1.31, 95% CI 1.12 to 1.53), MSMW (RR=1.47, 95% CI 1.26 to 1.72) and women (RR=2.25, 95% CI 2.03 to 2.49) (table 3). For MSM (RR=1.03, 95% CI 0.99 to 1.07), the relative risk remained similar post-COVID-19, whereas for transgender and gender diverse people (RR=1.27, 95% CI 0.90 to 1.79), the relative risk increased, although not statistically significant.

Factors associated with changes in Ng

The number of sexual partners and prevalence of inconsistent condom use, drug use during sex and history of STIs are visualised in online supplemental table 1. Among MSW, the likelihood of Ng significantly increased post-COVID-19 among those with 4–6 (p=0.019) and seven or more (p<0.001) sexual partners, who inconsistently used condoms (p=0.002), who did not use other drugs than alcohol and chems before or during sex (p=0.005), and those without a history of a bacterial STI (p=0.001) (table 4, online supplemental figure 2). Among MSMW, the likelihood of Ng increased post-COVID-19 among those with more sexual partners (4–6 sexual partners, p=0.008; 7 or more sexual partners, p<0.001), who inconsistently used condoms (p<0.001), and those without a history of a bacterial STI (p=0.029) (table 4, (online supplemental figure 3). Among women, the likelihood of Ng increased over time regardless of changes in sexual behaviour (table 4, online supplemental figure 4).

Table 4

Effect of behavioural determinants on changes in Ng positivity over time, Center for Sexual Health, Amsterdam, the Netherlands, 1 January 2018 to 31 December 2023

Ng-related symptoms

Pre-COVID-19, the prevalence of symptoms was 23.2% for MSM, 60.4% for MSW, 33.2% for MSMW, 17.1% for women and 18.0% for transgender and gender diverse people (online supplemental figure 5).

For MSM, the relative risk of having symptoms indicative for Ng was lower post-COVID-19 (RR=0.81, 95% CI 0.75 to 0.87) (online supplemental table 2). For MSW (RR=0.90, 95% CI 0.79 to 1.03), MSMW (RR=0.82, 95% CI 0.64 to 1.04) and women (RR=0.81, 95% CI 0.64 to 1.03), post-COVID-19, the relative risk was also lower, although not statistically significant. For transgender and gender diverse people, the relative risk pre-COVID-19 and post-COVID-19 was similar (RR=1.01, 95% CI 0.57 to 1.79).

Clinical complications of Ng

The incidence of PID among women and transgender and gender diverse people was 0.09% (42 diagnoses in 44 682 consultations) pre-COVID-19, and 0.09% (36 diagnoses in 40 077 consultations) post-COVID-19 (online supplemental table 3). The risk of PID did not change post-COVID-19 (RR=0.97, 95% CI 0.62 to 1.53).

The positivity rate of epididymitis among men and transgender and gender diverse people with a penis was 0.05% (33 diagnoses in 66 194 consultations) pre-COVID-19 and 0.04% (28 diagnoses in 68 195 consultations) post-COVID-19. The risk of epididymitis was lower post-COVID-19 (RR=0.80, 95% CI 0.48 to 1.35), although not statistically significant.

Discussion

We found an increased relative risk of Ng positivity post-COVID-19 among MSW, MSMW and women visiting the CSH in Amsterdam. In these groups, the increase was seen in all anatomical locations where Ng was tested. Among MSW and MSMW, Ng positivity significantly increased post-COVID-19 in those with a higher number of sexual partners, inconsistent condom use and no history of a bacterial STI. We did not find an increase in Ng among MSM and transgender and gender diverse people. Similarly, no increase in risk of PID or epididymitis was found

This concerning rise in Ng prevalence among MSW and women was also seen in other European countries.5 7–9 Data from Denmark also indicated that, along with an increase in Ng infections, there was a slight increase in chlamydia infections in 2023 compared with 2022.9 In addition, EU/EEA surveillance data showed an increase in chlamydia and syphilis diagnoses in 2022 regardless of gender.7 It is noteworthy that a rise in chlamydia diagnosis has not been seen in the Netherlands.3 We found that an increased number of sexual partners and inconsistent condom use were associated with an increased risk of Ng. Indeed, condom use has decreased particularly among younger individuals,10 which could translate to higher transmission of Ng. The decrease in condom use has also been reported among young people in the USA, Belgium and Spain.11–13 Noteworthy is the potential effect of the COVID-19 pandemic on the rise of Ng positivity. Due to restricted social interactions during the COVID-19 pandemic, young individuals were affected socially, which could have led to reduced sexual experimentation.14 Potentially, sexual behaviour, including the number of sexual partners, condom use and sexual networks, changed post-COVID-19 due to a catch-up effect. Although the transmission and epidemiological patterns differ from Ng infections, it is noteworthy that in the Netherlands, two recent emerging STI outbreaks occurred during-COVID-19 and/or post-COVID-19: scabies among young individuals and mpox among GBMSM.15 16 We also found an increase in Ng among MSMW and women during the COVID-19 pandemic. However, during that period, the CSH had a restrictive policy for seeing clients; as a result, fewer clients were tested for STIs, and those tested were a selected group with a higher risk of actually having an STI (because they reported symptoms, or were notified by a sexual partner). This leads to a higher positivity rate is to be expected. Therefore, we have not focused on the rise during the COVID-19 period.

Recent data indicate that gender and sexual fluidity increased among younger individuals in 2023 compared with 2017.10 As Ng was previously predominantly diagnosed among GBMSM in the Netherlands, a possible reason for the increase of Ng is MSMW acting as a bridging population between MSM and women (and indirectly MSW).7 17 18 A study in Australia showed that MSMW with Ng were dispersed within MSM and heterosexual networks, suggesting possible bridging resulting in transmission between two sexual networks.18 On the other hand, studies from Denmark, Spain and the Netherlands have shown distinct Ng strains in GBMSM and MSW/women, although MSMW were not analysed as a separate sexual network.9 19 20 One potential consequence of bridging between sexual networks is the introduction of Ng strains with lower antimicrobial susceptibility to heterosexual individuals, as Ng with lower susceptibility is more common among GBMSM.21 Additionally, recent sporadic cases of ceftriaxone-resistant Ng in Europe have been found in heterosexual individuals.22 However, ceftriaxone-resistant Ng strains have not yet been reported in the Netherlands,2 thus the risk of treatment failure is likely limited.

It is important to continue Ng surveillance for MSW, MSMW and women. Additionally, structural national campaigns regarding sexual health may need to be reintroduced. Focus should be on the prevention of STIs, sexual health education and improve STI testing accessibility for heterosexual and bisexual individuals. One option is to actively promote the use of condoms, for instance, on social media and at social gatherings. Providing condoms free-of-charge at pharmacies and STI clinics, as France has done since 2023 for individuals under 26 years old, may make access to condoms easier.23 Another approach is to increase STI screening and treatment for heterosexual and bisexual individuals. Since it is costly to expand testing capacity at STI clinics, home-based self-sampling can be offered if laboratories have sufficient capacity. Another option is to make STI tests publicly available, for example, through digital vending machines.24 This approach was recently tested in the United Kingdom and has proven to be an acceptable and effective to reach individuals who previously only tested for STIs infrequently or never.24 It is important to consider that increased testing will also result in more diagnoses of asymptomatic STIs that may resolve on their own.25 26 The increase of antibiotic consumption as a result could induce antimicrobial resistance.

Another consideration is the necessity to assess clinical complications, resulting from the increase of Ng infections. For example, a stark increase of Ng may lead to a higher incidence of PID in women, subsequently contributing to elevated rates of ectopic pregnancies and tubal factor infertility, especially since gonorrhoea is more likely to cause severe PID.27 Although we did not observe an increase of PID, it is important to note that women with acute symptoms suggestive of PID are mostly seen by general practitioners or hospitals. It is remarkable that the proportion of individuals with Ng, who also had symptoms, decreased. The authors of a recent study from Denmark hypothesised that the currently circulating Ng strains among MSW and women cause infection with no or low-grade symptoms but with a higher transmissibility.9 More research is needed to further clarify these findings.

This study is not without limitations. First, STI screening is freely available for MSW and women <25 years old. For MSW and women ≥25 years, screening is only available if they are symptomatic or have a partner notification. This could have introduced selection bias. Second, other factors, such as type of sexual network (eg, women having sexual contact with MSMW), could have played a role in the higher Ng incidence but were not routinely collected. Third, in view of the rising trends of genital Ng among women, oral positivity in MSW may also have risen. However, MSW are not test routinely for pharyngeal Ng, so we do not know what pharyngeal gonorrhoea positivity trends are among MSW. Testing MSW would be of interest from a surveillance perspective, but findings have no clinical consequences and will add substantial extra costs. Last, testing for Ng is also done outside of the CSH (eg, by general practitioners), but these data were not available for analysis. If the distribution of testing between CSHs and GPs has remained stable over the study period, and the characteristics of individuals testing outside the CSH as well, testing elsewhere should have no impact on our results. We have no reason to assume that such shifts have taken place, but we have no data to confirm this.

In conclusion, while the Ng positivity rate is generally higher among MSM and transgender and gender diverse people, we found that the risk of Ng increased over time for MSW, MSMW and women. Increased numbers of sexual partners and inconsistent condom use likely contributed to this rise. Given increasing gender and sexual fluidity, MSMW may serve as a bridging population between MSM and heterosexual individuals and increased transmission of Ng between these populations. Continuous surveillance, alongside promoting the importance of condoms and increasing STI services, may be crucial to curb the rise of Ng.