Article Text

Download PDFPDF

Shigella is common in symptomatic and asymptomatic men who have sex with men visiting a sexual health clinic in Amsterdam
Free
  1. Joyce F. Braam1,
  2. Sylvia M. Bruisten1,2,
  3. Mariska Hoogeland1,
  4. Henry J.C. de Vries1,3,
  5. Maarten F. Schim van der Loeff1,4,
  6. Alje P. van Dam1,2
  1. 1 Department of Infectious Diseases, Public Health Service of Amsterdam, Amsterdam, The Netherlands
  2. 2 Department of Medical Microbiology, Amsterdam UMC Locatie AMC, Amsterdam, North Holland, The Netherlands
  3. 3 Department of Dermatology, Amsterdam UMC Locatie AMC, Amsterdam, North Holland, The Netherlands
  4. 4 Department of Internal Medicine, Amsterdam UMC Locatie AMC, Amsterdam, North Holland, The Netherlands
  1. Correspondence to Joyce F. Braam, Department of Infectious Diseases, Public Health Service of Amsterdam, 1018 WT Amsterdam, The Netherlands; jbraam{at}ggd.amsterdam.nl

Abstract

Introduction Shigellosis is a reportable infectious disease. It can present as a severe bloody diarrhoea but is often asymptomatic. Shigella can be sexually transmissible. We performed a study among symptomatic and asymptomatic men who have sex with men (MSM) to assess the prevalence of Shigella, Salmonella and Campylobacter.

Methods From March to June 2020, MSM attending the Amsterdam centre for sexual health were consecutively included. Predefined minimal numbers of inclusion of 150, 100 and 50 were determined, respectively, for MSM who reported no diarrhoea, diarrhoea during last month or diarrhoea on the day of visit to clinic. Anal samples were tested for the presence of Shigella, Salmonella and Campylobacter. During the same period, the frequency of these bacteria was assessed in routinely tested samples requested by general physicians or nursing home physicians. Characteristics of included MSM were compared between the men with different diarrhoea anamnesis, and the prevalence of shigellosis was estimated in each group.

Results We included 212 MSM without diarrhoea, 109 MSM who recently had diarrhoea and 68 MSM who reported diarrhoea on the day of clinic visit. Thirteen (3.3%, 95% CI 1.7% to 5.6%) MSM were infected with Shigella, none with Salmonella and 7 (1.8%, 95% CI 0.7% to 3.7%) with Campylobacter. Shigella prevalence was 2.8% (95% CI 1.0% to 6.1%) in asymptomatic men, 3.7% (95% CI 1.0% to 9.1%) in men who recently had diarrhoea and 4.4% (95% CI 0.9% to 12.4%) in men with current diarrhoea (p=0.799). Shigella was more frequently found in MSM who had used pre-exposure prophylaxis (PrEP) in the preceding 3 months (10/151), compared with those not having used PrEP (2/146) or being HIV positive (1/75) (p=0.038). Shigella was significantly more often detected among MSM compared with routinely obtained faecal samples being 11/770 (1.4%) (p=0.031).

Conclusion Shigella infections are relatively common in both symptomatic and asymptomatic MSM. Future studies should focus on the risk of onward transmission via asymptomatic persons.

Samenvatting

Introductie

Shigellose is een meldingsplichtige infectieziekte. Het kan zich presenteren als een ernstige bloederige diarree, maar is vaak asymptomatisch. Shigella kan seksueel overdraagbaar zijn. We hebben een onderzoek uitgevoerd onder symptomatische en asymptomatische mannen die seks hebben met mannen (MSM) om de prevalentie van Shigella, Salmonella en Campylobacter te bepalen.

Methoden

Van maart tot juni 2020 werden achtereenvolgens MSM van het Amsterdamse centrum voor seksuele gezondheid opgenomen. Vooraf gedefinieerde minimale aantallen van inclusie van respectievelijk 150, 100 en 50 waren bepaald voor MSM die geen diarree, diarree in de afgelopen maand of diarree op de dag van bezoek aan de kliniek meldden. Anale monsters werden getest op de aanwezigheid van Shigella, Salmonella en Campylobacter. In dezelfde periode werd de frequentie van deze bacteriën bepaald in routinematig geteste monsters aangevraagd door huisartsen of verpleeghuisartsen. Kenmerken van geïncludeerde MSM werden vergeleken tussen mannen met verschillende diarree anamnese, en de prevalentie van shigellose werd in elke groep geschat.

Resultaten

We includeerden 212 MSM zonder diarree, 109 MSM die onlangs diarree hadden en 68 MSM die diarree meldden op de dag van het bezoek aan de kliniek. Dertien (3,3%, 95% CI 1,7-5,6%) MSM waren geïnfecteerd met Shigella, geen enkele met Salmonella, en 7 (1,8%, 95% CI 0,7-3,7%) met Campylobacter. De prevalentie van Shigella was 2,8% (95%CI 1,0-6,1%) bij asymptomatische mannen, 3,7% (95%CI 1,0-9,1%) bij mannen die recent diarree hadden en 4,4% (95%CI 0,9-12,4%) bij mannen met huidige diarree (P=0,799). Shigella werd vaker gevonden bij MSM die in de voorgaande drie maanden (10/151) PrEP hadden gebruikt dan bij mensen die geen PrEP hadden gebruikt (2/146) of hiv-positief waren (1/75) (p=0,038). Shigella werd significant vaker gedetecteerd bij MSM in vergelijking met routinematig verkregen fecale monsters, namelijk 11/770 (1,4%) (p=0,031).

Conclusie

Shigella infecties komen relatief vaak voor bij zowel symptomatische als asymptomatische MSM. Toekomstige studies moeten zich richten op het risico van verdere overdracht via asymptomatische personen.

  • homosexuality
  • male
  • homosexuality
  • male
  • bacterial infections
  • diagnosis

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information. Not applicable.

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Introduction

Shigella is an important cause of bacterial diarrhoea. Shigellosis, also known as bacillary dysentery, is a reportable infectious disease in the Netherlands and can present as a severe bloody diarrhoea.1 Not all infections result in clinical symptoms and not all persons with clinical symptoms are diagnosed.1 In Western countries, imported and domestically acquired shigellosis causes a substantial health-related burden such as disease control measures, use of healthcare facilities and a high number of disability adjusted life years.2 3

In many countries, including the Netherlands, isolation of Shigella is required for notification of shigellosis.4–6 The sensitivity of culturing is limited,7 but antimicrobial susceptibility testing is important, as Shigella species are developing resistance against several antibiotic groups.8 Shigella can be detected by molecular methods targeting the ipaH gene, but a distinction with entero-invasive Escherichia coli (EIEC) cannot be made with this technique. However, both Shigella and EIEC can cause serious disease,9 and public health consequences of EIEC and Shigella infections are similar.

About 22% of reported shigellosis cases in the Netherlands occur in men who have sex with men (MSM).9 In a Danish study, MSM had a matched OR (mOR) of 105 of Shigella infections compared with controls.10 In a US study on gastroenteritis in MSM, 30% had shigellosis.11 Several outbreaks of multidrug-resistant Shigella strains, such as with S. flexneri 3a in England, have been reported among MSM.12–14 Sexual behaviour such as oro-anal contact, condomless anal intercourse, multiple sexual partners, chemsex and encounters arranged through mobile dating applications are associated with shigellosis outbreaks among MSM.15

In the Netherlands in 2009–2016, 360–752 shigellosis cases yearly were confirmed and notified.2 The real number of cases is expected to be much higher. Haagsma et al estimated substantial underreporting, especially if diarrhoea is not bloody and caused by the limited sensitivity of culture; for every notified Shigella case, approximately 53 cases may go undiagnosed.16

Duration of persistence of Shigella after diarrhoea is generally unknown. Persistence of Shigella after 30 days in children in a low-income country was as high as 79%, but this could be related to persistent contamination of food or water.17 In adults in high-income countries, no systematic studies regarding duration of Shigella carriership or Shigella PCR positivity have been done, although Shigella long-term carrier state has been described.18 So possibly in a population with a high frequency of Shigellosis, such as MSM, Shigella may be regularly detected in stool during but also shortly after an episode of diarrhoea.1

Asymptomatic infections with Shigella seem to be uncommon in the general population. In a previous Dutch study, Shigella was detected in 0.9% of patients with diarrhoea (n=1515) but not at all in healthy controls (n=1195).19 However, this study did not focus on key populations like MSM. Little is known whether and how often asymptomatic Shigella infections occur in MSM.20

Salmonella and Campylobacter have also been reported among MSM and to be sexually transmissible.21 22 The most common detected bacteria in stool samples from MSM with gastroenteritis in the USA were diarrheagenic Escherichia coli (33%), Shigella (30%) and Campylobacter (17%).11 In a Danish study, MSM had a mOR of 16 for Campylobacter infection compared with controls, but no significantly increased mOR was found for Salmonella infection.10 Since Salmonella and Campylobacter are not notifiable pathogens in The Netherlands, it is unknown what the prevalence or incidence of these infections are among MSM.

Targeted research into pathogens in MSM with or without diarrhoea has not been conducted in Amsterdam before. Therefore, we performed a study among MSM visiting the STI clinic in Amsterdam to assess the prevalence of Shigella, Salmonella and Campylobacter.

Methods

Study design and inclusion criteria

This prospective study included MSM ≥16 years old attending the STI clinic in Amsterdam in March–June 2020. We aimed to include samples of 150 asymptomatic men, 100 men with diarrhoea in the previous month and 50 with diarrhoea on the day of clinic visit (see further). Men from the different diarrhoea categories were consecutively included, until the target number for that category was reached. Diarrhoea was defined as thin or watery stools, or bowel passages more than two times daily. Age, ethnicity, sexual behaviour, STI diagnoses, HIV status, pre-exposure prophylaxis (PrEP) use and the presence of clinical symptoms were extracted from electronic patient files.

Study sample size

As a reference, we used the prevalence of 0/1195 (0%, 95% CI 0% to 0.3%) Shigella infections in healthy controls in the general population in another Dutch study.19 We assumed a prevalence of 2% Shigella among asymptomatic MSM. As a reference for the presence of shigellosis in MSM with diarrhoea, we used 30% as found in a study conducted in the USA11 and divided this by two since we expect a lower prevalence in the Netherlands, resulting in an expected prevalence of 15% among MSM with diarrhoea. In order to determine a statistically significant difference between the prevalence of Shigella in asymptomatic MSM and MSM with current diarrhoea, we would have to include 150 and 50 persons, respectively. We estimated that the Shigella prevalence would decline by 50% in the first month after diarrhoea and therefore aimed to include 100 MSM with recent diarrhoea.

Detection of bacterial pathogens

Anal swab samples were routinely taken to detect Chlamydia trachomatis and Neisseria gonorrhoeae with the Aptima Combo 2 (AC2) transcription-mediated amplification assay (Hologic Inc, San Diego, California, USA) according to the manufacturer’s instructions. Use of routinely taken anal swabs to detect C. trachomatis and next to detect gastrointestinal infections has been used before.20 DNA was extracted with the MagNA Pure 24 System (Roche, Rotkreuz, Switzerland) and tested for the presence of Shigella/EIEC by qPCR on the ipaH gene, Salmonella on the ttrC gene and Campylobacter on the mapA and ceuE genes. Primers and probes are listed in table 1. Amplification was performed on a RotorGene (Qiagen, Hilden, Germany) with the following program for Shigella, Salmonella and Campylobacter: 2 min at 50°C, 2 min at 95°C followed by 45 cycles of 95°C for 15 s and 60°C for 45 s. Since clinical symptoms and public health consequences of Shigella and EIEC are similar,9 positive ipaH qPCR results are reported as Shigella infections throughout the remainder of the manuscript. A sample was considered positive for Campylobacter if one or both of the genes were positive with qPCR. Shigella, Salmonella and Campylobacter testing were done for study purposes only, and results were neither disclosed to the men nor to the healthcare professionals. During the study period, the frequency of Shigella, Salmonella and Campylobacter was assessed by qPCR in routinely tested samples requested by general physicians or nursing home physicians. For them, the main and almost only reason to send a faecal sample for testing for Shigella, Salmonella and Campylobacter is diarrhoea. No specific clinical information about symptoms was available from these routinely tested samples. Only shigellosis is a notifiable disease in the Netherlands. We collected data regarding the number of notified cases and distributions among different categories in the Amsterdam region.

Table 1

Primers and probes for Shigella/EIEC, Salmonella and Campylobacter

Statistical analysis

Characteristics of included MSM clients were compared between the men with different diarrhoea anamnesis, and the prevalence of shigellosis was estimated in each group. Also, prevalence was estimated and compared between groups of demographic and sexual behavioural variables, with 95% CIs. Differences were assessed with the χ2 test or Fisher’s exact test, except for age, which was compared using the Kruskal-Wall test. Univariable logistic regression analysis was performed to identify possible risk factors for shigellosis. The prevalence of shigellosis was compared between MSM and routinely tested samples requested by general physicians or nursing home physicians in the study period, using the χ2 test. Level of significance was set at p<0.05. Data were analysed using Stata Intercooled 15 (StataCorp LLC, Texas, USA).

Ethics statement

Men of the STI outpatient clinic Amsterdam were informed of the ‘opt-out’ system regarding research on remnants of patient material. All data were pseudonymised before analysis.

Results

Study population characteristics

During the study period 2927 MSM clients visited the clinic, of whom 2307 (78.8%) reported no diarrhoea, 512 (17.5%) reported diarrhoea in the month before the visit and 110 (3.7%) reported diarrhoea on the day of the visit. We included samples from 212 MSM without diarrhoea, who were included 3–9 March 2020, 109 MSM who had diarrhoea in the preceding month included 3–16 March 2020 and 68 MSM who reported diarrhoea on the day of clinic visit, included 3 March–23 June 2020 (table 2). Of note, nationwide lockdown measures to inhibit SARS-CoV-2 transmission started on 16 March. Thus, most of the samples from MSM with diarrhoea were collected during the lockdown period. Median age was 32 years (IQR 26–44), 41.4% (161) were MSM of Dutch origin, 38.8% (151) reported PrEP use, 19.3% (75) were HIV-positive and 7.5% (29) reported anal symptoms, other than diarrhoea.

Table 2

Characteristics of MSM clients visiting the STI clinic Amsterdam, March–June 2020, who were included in the study, by diarrhoea history

Shigella infections

The total number of MSM infected by Shigella was 13/389 (3.3%; 95% CI 1.7% to 5.6%). Of MSM without diarrhoea 6/212 (2.8%; 95% CI 1.0% to 6.1%) were infected, 4/109 (3.7%; 95% CI 1.0% to 9.1%) of MSM who recently had diarrhoea and 3/68 (4.4%; 95% CI 0.9% to 12.4%) of MSM who had diarrhoea on the day of clinic visit (p=0.799) (table 2). During the study period, 11/770 (1.4%; 95% CI 0.7% to 2.5%) of routinely tested faecal samples sent by general physicians or nursing home physicians were positive for Shigella, significantly lower in comparison with MSM included in this study (p=0.031). Shigella positive samples were more frequently found in HIV negative MSM using or recently having used PrEP (10/151, 6.6%), compared with HIV negative MSM not using PrEP (2/146, 1.4%) or HIV positive MSM (1/75, 1.3%) (p=0.038) (table 3). No association was found between number of sex partners (0–1 vs 2 or more) in last 6 months, age, country of origin or coinfections. Only 3/13 (23.1%) of the men with Shigella had diarrhoea; one of them also had anal discharge and was diagnosed with syphilis. The other 10 men with a Shigella infection reported no symptoms. Three of 13 MSM with Shigella (23.1%) had a coinfection with C. trachomatis, one (7.7%) with N. gonorrhoeae and one (7.7%) with Treponema pallidum. Coinfections with hepatitis B, Campylobacter, Salmonella or SARS-CoV-2 were not observed. The number of notified Shigella infections among the entire population of Amsterdam region was 48 in 2020, of which 23 were from MSM, 15 were travel related and for 10 none of these applied.

Table 3

Associations between possible risk factors and Shigella infections among MSM routinely visiting the STI clinic Amsterdam, March–June 2020

Salmonella and Campylobacter infections

We detected no Salmonella infections in the included MSM population. In comparison, in routinely tested faecal samples during the same study period, 3/771 (0.4%; 95% CI 0.1 to 1.1%) samples were positive for Salmonella (p=0.555). From the MSM included in the study, 7/389 (1.8%; 95% CI 0.7% to 3.7%) were infected by Campylobacter. Infection with Campylobacter was 0.9% (2/212) among MSM without diarrhoea, 3.7% (4/109) among MSM reported to have had recently diarrhoea, and 1.5% (1/68) among MSM with diarrhoea at clinic visit. In comparison to all MSM included in this study, 19/770 (2.5%; 95% CI 1.5% to 3.8%) of routinely tested faecal samples during the study period were positive for Campylobacter (p=0.468).

Discussion

We found a positivity rate of 3.3% Shigella and 1.8% Campylobacter among MSM attending the STI clinic in Amsterdam in the spring of 2020. We did not detect any Salmonella infection during the study period. Shigella infections were significantly more prevalent among MSM compared with routinely tested samples requested by general physicians or nursing homes, and these routinely tested samples were most likely from symptomatic patients. No difference regarding Campylobacter infections was found. In this study population, none of the pathogens were significantly associated with diarrhoea at clinic visit or in the last month.

No significant difference in Shigella prevalence was found between MSM with current or recent diarrhoea compared with MSM without diarrhoea. This could well be due to a lack of power because of the small sample size. However, our findings suggest that shigellosis in MSM is frequently asymptomatic. Shigella-positive samples were more frequently found in HIV-negative MSM with recent PrEP use, compared with HIV-negative MSM not using PrEP or HIV positive MSM. This might be due to an higher number of sexual partners among PrEP users (115/151, 76.8% had five or more sex partners in the last 6 months) compared with no PrEP users (73/146, 50.0% had five or more sex partners in the last 6 months), potentially including a higher frequency of oro-anal contacts. However, many HIV-positive MSM similarly had five or more sex partners in the last 6 months (60/75, 80.0%). We did not find a correlation between Shigella infection and the number of sexual partners. However, we discriminated only between 0–1 and 2 or more sexual partners in the last 6 months. The number of Shigella infections was too low to perform a more detailed analysis. Previously oro-anal contact was associated with transmission of C. trachomatis, N. gonorrhoeae 23 and extended-spectrum β-lactamase Enterobacteriaceae.24 Oro-anal contact seems to be the main risk factor for acquiring Shigella infections in MSM.

Extrapolating from the prevalences of 2.8%, 3.7% and 4.4% of shigellosis among 212 asymptomatic, 109 recently symptomatic and 68 symptomatic MSM in our study population to the 2307 asymptomatic, 512 recently symptomatic and 110 symptomatic MSM that visited the clinic during the study period, it can be expected that during the 16-week study period 65 asymptomatic (95% CI 24 to 140), 19 (95% CI 5 to 47) recently symptomatic and 5 (95% CI 1 to 14) symptomatic MSM were infected with Shigella. Thus, in total, 89 men (95% CI 30 to 201) out of the 2929 MSM clients attending the STI clinic in a 16-week period might have been infected with Shigella. Considering that in 2020 in Amsterdam region 48 cases of shigellosis were notified, of which 23 were from MSM, this suggest that the actual prevalence of shigellosis is much higher than the number of notified cases. A positive culture is mandatory for the notification and cultures often remain negative, due to low load or to the presence of EIEC causing the positive PCR result.9

Previously, Shigella was detected in 0.9% of patients with diarrhoea in the Netherlands but not detected in healthy controls.19 In contrast, we found that 2.8% of the MSM without diarrhoea were infected with Shigella, which was not significantly different compared with MSM with diarrhoea or who had diarrhoea in the previous month. Thus, asymptomatic Shigella infections do occur relatively frequently in MSM. The impact of diagnosing asymptomatic Shigella infections depends highly on the frequency that they will cause secondary, symptomatic infections. This should be further investigated.

In our study, the prevalence of Campylobacter and Salmonella infections were lower in MSM with or without diarrhoea compared with what was previously reported.19 In a Danish study, Campylobacter and Shigella infections were associated with being MSM, but Salmonella infection was not.10 Whether or not sexual transmission plays a significant role in Campylobacter and Salmonella infections is still unclear.

A limitation of this study is that because of the start of the SARS-CoV-2 epidemic in the Netherlands, our results might not be fully translatable to a time without an epidemic. Since the national lockdown on 16 March 2020, international travel was highly restricted, bars and restaurants were closed and advice was given against social encounters. This lockdown lasted until after the end of our study period, although some restrictions were eased from 1 June 2020, onwards. Access to the STI clinic in that period was highly restricted and only persons with clinical symptoms and persons at high risk of STIs were seen at the clinic. All samples from persons without diarrhoea and the majority of persons with recent diarrhoea had already been collected before the onset of the lockdown. A large proportion of samples from patients with diarrhoea at clinic visit, however, were collected during the lockdown period, during which the epidemiology of infectious diseases was possibly changed and less Shigella may have circulated. Another limitation of this study is that we could not make a distinction between Shigella and EIEC infections. A strength of this study is that we did not only take into account diarrhoea at time of visit, but also diarrhoea in the previous month.

To conclude, Shigella is prevalent among both symptomatic and asymptomatic MSM. Whether and when MSM should be tested for Shigella in MSM to prevent the spread of this notifiable pathogen remains unclear. Future studies should focus on the risk of Shigella transmission of asymptomatic MSM, possibly leading to symptomatic infections in their partners.

Key messages

  • Men who have sex with men (MSM) have Shigella infections more often than the general population.

  • In the Netherlands, prevalence of Shigella among symptomatic (4.4%) and also asymptomatic (2.8%) MSM is high.

  • Shigella is more frequently found in MSM who used pre-exposure prophylaxis (PrEP), compared with those without a history of PrEP use and HIV-positive men.

  • Salmonella and Campylobacter were not found more frequently in MSM than the general population.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information. Not applicable.

Ethics statements

Patient consent for publication

Ethics approval

This study involves human participants, but Medical Ethics Committee of the University Amsterdam Medical Centers W20_151 # 20.167 exempted this study. Men of the STI outpatient clinic Amsterdam were informed of the 'opt-out' system regarding research on remnants of patient material. All data were pseudonymised before analysis.

Acknowledgments

The authors would like to thank Eline van Dulm, Lian Bovée and Mariska Petrignani of the Public Health Service of Amsterdam for their input in the study design.

References

Footnotes

  • Handling editor Tristan J Barber

  • Twitter @henryjdevries

  • Contributors SMB, HJCdV, MSvdL and AvD contributed to the design and implementation of the research; MH performed the measurements; JB performed the analysis of the results with input from MSvdL and AvD; and JB wrote the paper with input from all authors; AvD acted as guarantor.

  • Funding Funding was provided by the Public Health Laboratory of GGD Amsterdam.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.