Article Text

Original article
Is rectal douching and sharing douching equipment associated with anorectal chlamydia and gonorrhoea? A cross-sectional study among men who have sex with men
  1. RCA Achterbergh1,
  2. JJ van der Helm1,
  3. W van den Boom1,
  4. T Heijman1,
  5. IG Stolte1,
  6. MS van Rooijen1,
  7. HJC de Vries1,2,3
  1. 1 STI outpatient clinic, department of infectious diseases, Public Health Service Amsterdam, Amsterdam, The Netherlands
  2. 2 Department of Dermatology, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands
  3. 3 Centre for Infection and Immunity Amsterdam (CINIMA), Academic Medical Centre, University of Amsterdam, Amsterdam, The Netherlands
  1. Correspondence to Professor HJC de Vries, STI Outpatient Clinic, Department of Infectious Diseases, Public Health Service Amsterdam, Nieuwe Achtergracht 100, Amsterdam 1018 WT, The Netherlands; h.j.devries{at}amc.uva.nl

Abstract

Introduction Men who have sex with men (MSM) are at high risk for anorectal chlamydia and gonorrhoea infections. Many MSM use rectal douches in preparation for sex, which might break down the mucosal barrier function and facilitate the acquisition of STI. We determined whether rectal douching or sharing douching equipment was associated with anorectal chlamydia and gonorrhoea.

Methods In a cross-sectional study among 994 MSM attending the STI outpatient clinic of Amsterdam between February and April 2011, data were collected on rectal douching, sexual behaviour and STI. We used multivariable logistic regression analysis to determine the association between rectal douching, including sharing of douching equipment, and anorectal chlamydia and gonorrhoea for those reporting receptive anal sex. We adjusted for other risk behaviour, that is, condom use, number of partners and HIV status.

Results Of 994 MSM, 46% (n=460) practised rectal douching, of whom 25% (n=117) shared douching equipment. Median age was 39 years (IQR 30–47), median number of sex partners in the 6 months prior to consult was five (IQR 3–10) and 289 (29.0%) participants were HIV positive. The prevalence of anorectal chlamydia and/or gonorrhoea for those reporting receptive anal sex was 9.6% (n=96). In multivariable analysis, HIV positivity (aOR=2.2, 95% CI 1.3 to 3.6), younger age (aOR=2.5, CI 1.4 to 4.5 for those aged <35 years compared with those aged ≥45 years), and more sexual partners (aOR=1.2, 95% CI 1.0 to 1.5 for 1 log increase) were significantly associated with anorectal STI. However, rectal douching or sharing douching equipment were not significantly associated with anorectal chlamydia and/or gonorrhoea (p=0.647).

Conclusions Almost half of MSM used rectal douching and a quarter of these shared douching equipment. Though using douching equipment does not appear to contribute to anorectal chlamydia and gonorrhoea in this study, STI prevalence remains high and prevention strategies like early testing and treatment remain of utmost importance.

  • BACTERIAL INFECTION
  • GAY MEN
  • PREVENTION
  • PUBLIC HEALTH
  • SEXUAL BEHAVIOUR

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

Men who have sex with men (MSM) are at high risk for anorectal STI; 10.3% of the MSM attending the STI outpatient clinic of Amsterdam tested positive for anorectal chlamydia and/or gonorrhoea in 2011.1 Many MSM use rectal douches in preparation for sex,2–6 mainly for hygienic reasons.3 ,4 It has been suggested that rectal douching breaks down the mucosal barrier function and facilitates the acquisition of STI pathogens.7 ,8 If shared, contaminated douche equipment itself could also be a transmission route. Thus, if douching or sharing douching equipment were to be associated with increased anorectal chlamydia and gonorrhoea prevalence, public health messages targeted at MSM should be adapted, and rectal douching or sharing equipment should be discouraged.

Several studies have shown higher STI prevalence among people who use rectal douches.3 ,4 ,8–11 These studies, however, were limited by their design as they did not consider the anatomical location—urethral, pharyngeal or anorectal—of the STI, except for two case-control studies that only assessed Lymphogranuloma venereum (LGV).8 ,11 To our knowledge, studies that assessed the association of rectal douching and anorectal, laboratory-confirmed chlamydia and/or gonorrhoea have not been performed to date.

In this cross-sectional study among MSM attending the STI outpatient clinic of Amsterdam, we elucidated the following questions: (1) Which proportion of MSM use rectal douches and/or share douche equipment with sex partners and what are characteristics of douching behaviour? (2) What are characteristics of MSM who use anorectal douches? (3) Are rectal douching and/or sharing douche equipment associated with anorectal chlamydia (including LGV) and anorectal gonorrhoea?

Methods

Study population

MSM were recruited at the STI outpatient clinic of the Public Health Service of Amsterdam, which is a low-threshold clinic annually serving approximately 30 000 visitors.1 Individuals can attend the STI clinic anonymously, free of charge and without a referral by a medical doctor. Individuals are prioritised based on a short questionnaire to estimate risk of having an STI.12 MSM are considered high-risk clients and get a full STI check, including physical examination. Because routinely collected anonymous data was used for this study, no ethical clearance was sought.

Demographic characteristics, rectal douching and sexual behaviour

All men, 18 years and older, who reported having sex with another men in the preceding 6 months attending the STI outpatient clinic of Amsterdam between 14 February 2011 and 6 April 2011 were included in the study once. Trained healthcare workers collected routine data in an electronic patient database during consultations as described previously.12 Additional data on rectal douching behaviour were collected during consultations by healthcare workers comprising whether they used rectal douches in the preceding 6 months (yes, no), and if so, the frequency of douching, the moment of douching (before anal intercourse, after anal intercourse, before using toys, before fisting, before rimming or other), type of douching equipment (showerhead hose without nozzle, showerhead hose with nozzle, enema, syringe, enema bottle or bulb or other), whether they shared their douching equipment (with steady partner, with known casual partner, with unknown casual partner, with ex-steady partner, public sex location, never share douching equipment) and location of douching (at home, at sex partner's home, sauna, sex club, hotel or other).

Routinely collected data also included sex of partner (women, men), type of partner ((1) steady partner (2) known casual/ex-steady partner, (3) casual partner or (4) anonymous/sex worker/client), condom use (condom, no condom—including condom failure) and position during anal sex (no anal sex, insertive only, receptive only or insertive and receptive) during the preceding 6 months.

Specimen collection and testing procedures

Medical staff obtained pharyngeal and anorectal swabs. Urine samples, pharyngeal and anorectal swabs were tested for gonorrhoea and chlamydia using the Aptima combo-2 assay (Gen-probe Incorporated, San Diego, California, USA).

Serum testing for syphilis was performed with the treponema pallidum particle agglutination assay (Fujirebio, Tokyo, Japan); if reactive, the rapid plasma reagin card test (RPR nosticon II, Biomérieux, Marcy l'Etoile, France) and the fluorescent treponema antibody absorption test (Trepo-spot IF, Biomérieux, Marcy l'Etoile, France) were performed to confirm and classify the stage of infection.

Unless they opted out, all participants who had not previously tested HIV positive were tested for HIV using a rapid immunoassay (Alere Determine HIV1/2; Alere Health BV, Tilburg, the Netherlands). Reactive or indeterminate samples were confirmed by HIV Ag/Ab combo test (Axsym; Abbott Laboratories, Illinois, USA) and line immune assay (Inno-Lia HIV I/II Score; Innogenetics, Ghent, Belgium). If these confirmations were indeterminate or negative, a Vidas P24-antigen test (Biomerieux, Marcy l'Etoile, France) was performed.

Statistical analysis

An anonymous database was assembled and analysed with SPSS package V.21.0 (SPSS, Chicago, Illinois, USA). Our aim was to include 1000 participants.

The χ2 test was used to examine whether epidemiological characteristics and STI prevalence differed between MSM who reported rectal douching and MSM who did not. We considered p<0.05 as statistically significant.

To assess determinants for the use of rectal douches, we performed logistic regression analysis including the following variables: age, ethnicity, condom use, number of partners in the preceding 6 months, HIV status, type of partner (steady partner, known casual/ex-steady partner, casual partner or anonymous/sex worker/client) and sexual position (no anal sex, insertive, receptive or insertive and receptive). Variables that were associated with rectal douching at p≤0.10 in univariable analysis were entered into a multivariable logistic regression model using backward selection.

To assess determinants of anorectal chlamydia (both non-LGV and LGV) and/or gonorrhoea, we performed univariable logistic regression analysis including the following variables: age, ethnicity, condom use, number of partners in the preceding 6 months, HIV status, most high-risk type of partner, douching and sharing douching equipment. Only those at risk for anorectal chlamydia and/or gonorrhoea, that is, MSM who reported receptive anal intercourse were included in the analysis. Age was divided into three categories based on tertiles and international standards. We used the logarithm of the number of sexual partners in the multivariable analysis to better normalise the distribution. We grouped a small number of participants with a non-Dutch ethnicity into a non-Dutch category. Variables that were associated with anorectal chlamydia and/or gonorrhoea at p≤0.10 in univariable analysis were entered into a multivariable logistic regression model using backward selection. Number of sex partners, HIV status, condomless anal sex (CAS) and age have been associated with anorectal chlamydia and/or gonorrhoea by other studies13–16 and thus were forced into the multivariable model. We checked for interactions between anorectal STI, HIV status and douching.

In sensitivity analysis, we divided douching into two categories; those who practiced anorectal douching and those who did not; we used anorectal chlamydia, gonorrhoea and/or syphilis as outcome. As receptive anal intercourse is a risk factor for syphilis,17 we hypothesised that anal douching might influence both anorectal as well as systemic bacterial STI.

Results

Study population

In total, 1028 consecutive MSM were tested for STI between 14 February 2011 and 6 April 2011 at the STI outpatient clinic of Amsterdam. In 21 consultations, the questionnaire on rectal douching was not filled out and those were excluded from the analysis. These MSM did not differ significantly in age, HIV status, number of partners or STI positivity compared with those included in the analysis. Thirteen MSM were included twice and their latter consult was excluded from the analysis. In total, 994 men were included in the study.

Demographic characteristics, sexual risk behaviour and STI data are shown in table 1.

Table 1

Demographic and behavioural characteristics and STI of men who have sex with men attending the STI outpatient clinic of Amsterdam, February to April 2011, according to rectal douching habits

The median age was 39 years (IQR 30–47). In total, 701 (70.5%) were of Dutch ethnicity. The median number of sex partners in the 6 months prior to consult was five (IQR 3–10) and 289 (29.0%) participants were HIV positive. Twenty (2.0%) MSM reported having had sex in exchange for money or goods. In total, 460 (46.3%) MSM reported rectal douching in the 6 months before attending the STI clinic.

STI prevalence

An STI was diagnosed at any anatomical location (chlamydia, gonorrhoea, LGV, syphilis, HIV, hepatitis C (HCV)) in 222 (22.3%) MSM. Sixty-one (6.2%) MSM were diagnosed with an anorectal chlamydia non-LGV infection, 4 (0.4%) with anorectal LGV and 47 (4.7%) with anorectal gonorrhoea. Nine men had an anorectal chlamydia and gonorrhoea coinfection. In total, 14 new HIV and 30 syphilis infections were diagnosed. MSM who reported rectal douching were more often infected with anorectal non-LGV chlamydia (p<0.001), anorectal gonorrhoea (p=0.001), syphilis (p=0.003), LGV (p=0.031) and newly diagnosed HIV (p=0.004). However, the number of chlamydia, non-LGV and gonorrhoea infections at any location—including urethral and pharyngeal—did not significantly differ between those who douched and those who did not (p=0.116 and p=0.081, respectively) (table 1).

Rectal douching

Of the 460 (46.3%) MSM who reported rectal douching, 58.3% reported douching ≥3 times a month (table 2).

Table 2

Rectal douching behaviour of 460 men who have sex with men by HIV status, attending the STI outpatient clinic of Amsterdam, February to April 2011

Most participants, 406 (88.6%), reported douching before receptive anal contact, whereas 59 (12.9%) MSM reported rectal douching after receptive anal contact and 20 (4.3%) reported other moments that were unrelated to sex, mainly after toilet visits. Men mainly used a showerhead hose without a nozzle (41.5%) or a showerhead hose with a nozzle (45.4%). Most men (74.9%) did not share their douching equipment, though HIV-positive men were more likely to share their douching equipment compared with HIV-negative men (30.8% vs 20.8%, respectively; p=0.027). Nearly all men (97.2%) reported douching at home, 50 (22.9%) of the HIV-positive men also reported douching at the home of their sex partner, which was significantly more than for HIV-negative men (9.6%) (p<0.001). Fourteen (3.1%) men had douched at public places (ie, in a darkroom, sauna or sex club).

Determinants for rectal douching

In a multivariable analysis among all 994 participants, several variables were significantly associated with MSM who reported anal douching. Those variables were older age, higher log number of sex partners, HIV positivity and receptive anal intercourse (table 3).

Table 3

Univariable and multivariable analyses of determinants for using rectal douches among 994 men who have sex with men attending the STI outpatient clinic of Amsterdam, February to April 2011

Determinants for anorectal chlamydia and/or gonorrhoea

In the univariable and multivariable analyses on anorectal chlamydia and/or gonorrhoea, only those at risk were included (ie, 679 men who reported receptive anal intercourse). In univariable analyses, non-Dutch ethnicity, a higher log number of sex partners, HIV status and CAS were significantly associated with anorectal chlamydia and/or gonorrhoea (table 4).

Table 4

Univariable and multivariable analysis of determinants for anorectal chlamydia, gonorrhoea and/or Lymphogranuloma venereum (n=96) among 679 men who have sex with men who reported receptive anal sex attending the STI outpatient clinic of Amsterdam, February to April 2011

Both rectal douching and sharing douching equipment showed a trend towards higher anorectal STI positivity; however, like age and type of partner, these variables were not significantly associated with anorectal chlamydia and/or gonorrhoea.

In multivariable analysis, younger age (aOR=2.6; 95% CI 1.4 to 4.6 for those aged <35 years and aOR=1.5 95% CI 0.8 to 2.8 for those aged 35–44 years compared with those aged ≥45 years), HIV positivity (aOR=2.2; 95% CI 1.3 to 3.6) and log number of sex partners (aOR=1.2; 95% CI 1.0 to 1.5 for 1 log increase) were significantly associated with anorectal chlamydia and/or gonorrhoea, whereas anorectal douching was not significantly associated, irrespective of whether douching equipment was shared or not (aOR 1.2; 95% CI 0.7 to 2.0 and aOR 1.4; 95% CI 0.7 to 2.8, respectively, compared with not douching p=0.647). We did not find an interaction between rectal douching and HIV status for anorectal STI.

We performed a sensitivity analysis comparing the effect of rectal douching irrespective of sharing douching equipment, using anorectal chlamydia, gonorrhoea and/or syphilis combined and separate as outcomes. These models showed similar results in multivariable analysis; no significant association was found between rectal douching and anorectal chlamydia, gonorrhoea and/or syphilis. Multivariable analysis stratified for HIV status showed similar trends (data not shown). Multivariable analysis for chlamydia and gonorrhoea separately showed no significant association (see online supplementary appendix).

Discussion

This study shows that 46% of MSM visiting the STI outpatient clinic in Amsterdam practised rectal douching in the preceding 6 months, and a quarter of these shared douching equipment. In multivariable analysis, both rectal douching with and without sharing douching equipment were not significantly associated with anorectal bacterial chlamydia and/or gonorrhoea. To our knowledge, this is the first study that describes the association between sharing douching equipment with partners and having a laboratory-confirmed chlamydia and/or gonorrhoea infection.

A strength of this study is that laboratory-confirmed STI diagnosis was used as outcome compared with most studies that use self-reported STI. In addition, adding syphilis as an outcome did not change our results, which strengthens our findings. A limitation is that our routine data are clinic-based rather than primarily designed to systematically collect in-depth data on specific determinants. Therefore, relevant data such as risk behaviour of all partners may have been missed, as we focused only on the highest risk encounters. We also did not collect data on other high-risk behaviours like fisting, using (sex) drugs or sharing toys18 nor the time interval between usage and sharing douching equipment. Not being able to control for other high-risk behaviour may result in an overestimation of the effect of anal douching. Our behavioural data were collected by nurses in an STI outpatient setting, making biased socially acceptable answers from respondents possible. It should also be noted that the MSM in our study are at increased risk for STI and do not represent the general MSM population. Among participants of the Amsterdam Cohort Studies (a longitudinal cohort of asymptomatic MSM) for example, the STI prevalence is much lower (5.9%).19

The prevalence of rectal douching among men in our study is comparable with previously reported results. In a Peruvian study among 824 MSM, 27% reported a history of rectal douching through a computer-assisted self-interview.2 In two large self-reported internet surveys, 35%–66% reported rectal douching.3 ,4 In these two surveys, MSM who used rectal douches were more often HIV positive, had more sex partners, had more CAS and more often identified as both receptive and insertive than those who did not use rectal douches. These findings were similar to ours; MSM who used douching equipment showed higher risk behaviour than MSM who do not use douching equipment.

Previous studies have demonstrated that douching practices by MSM are associated with an increased prevalence of STI like chlamydia,10 gonorrhoea,9 LGV8 ,11 and self-reported STI.3 ,4 Differences in the design of previous work and this study could explain the contrast between findings. Except for the case control study by Macdonald et al, no other study has included both the anatomical location of the STI3 ,4 ,9 ,10 and sexual position.3 ,4 ,8–10 When we performed a multivariable analysis with all 994 participants and omitted the variable sexual position, we found similar results to other studies—a significant association between douching and anorectal STI (data not shown). However, we included only those MSM exposed through receptive anal contact, as these men are at highest risk for anorectal infections and the association between douching and anorectal STI was not significant.

In the case-control study by Macdonald et al,11 an association was found between rectal LGV and rectal douching. Only 7% of the cases compared with a quarter of the controls reported any insertive anal practices, this potential bias might explain why douching persisted in their final model. Also, the study had limited scope to investigate HIV status as many men were HIV positive. Among HIV-positive MSM, the mucosal barrier is already negatively affected by HIV infection.20 The mucosal damage caused by douching could possibly further increase the risk of LGV acquisition in HIV-positive individuals. We did not find an interaction between HIV status, douching and anorectal STI.

Schmelzer et al 21 first demonstrated that large volume enemas could damage the rectal mucosa. Rectal biopsies showed surface epithelium loss after soapsuds and tap water but not after polyethylene glycol electrolyte solution enemas. Tap water or soapsuds are used by 77%–92% of the MSM reporting rectal douching.3 ,4 A previous study at the STI clinic of Amsterdam revealed that 66 out of 68 MSM used only tap water.8

In the case-control study by Macdonald et al,11 44% of the LGV-positive cases and 23% of the LGV-negative controls with potential LGV symptoms shared douching equipment. The prevalence of men who shared douching equipment in our study was 11.7%. Although not significant, we did find a trend towards higher anorectal chlamydia and gonorrhoea positivity among MSM who shared douching equipment compared with those who did not share douching equipment or those who did not use rectal douches. Since this trend is found in both uni- and multivariable analysis we cannot exclude that the lack of significance results from a power issue. Due to the relatively low sample size, we cannot draw definite conclusion about the validity of the observed non-significant association between rectal douching and anorectal STI; the power was 0.4. However, another explanation might be that MSM who share rectal douching equipment more often exhibit other risky behaviour—like sharing toys, using sex-related drugs or group sex—contributing to the increased STI positivity,18 which are possible unmeasured confounders.

In conclusion, we did not find an association between the use of rectal douches and anorectal chlamydia and/or gonorrhoea. Therefore, we believe that douching and sharing douching equipment cannot have a large contribution to anorectal chlamydia and gonorrhoea in MSM. This is of importance for the development of rectal microbicides. Although douching seems not associated with STI, STI prevalence among MSM practising douching is alarmingly high. Therefore, assessing sexual practices and techniques and promoting prevention strategies like condom use, early testing and treatment remain of utmost importance.

Key messages

  • Among men who have sex with men (MSM) attending the STI clinic of Amsterdam, 46% practised rectal douching in the preceding 6 months and a quarter of these shared douching equipment.

  • In this study, STI prevalence among MSM practising rectal douching was 26.7%.

  • In multivariable analysis, we did not find an association between the use of rectal douches and anorectal chlamydia and/or gonorrhoea.

Acknowledgments

We thank the patients who contributed to this study, and acknowledge the staff at the STI outpatient clinic of the Public Health Service of Amsterdam. We would like to thank Amy Matser for her contribution.

References

Supplementary materials

  • Abstract in Dutch

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

Footnotes

  • Handling editor Jackie A Cassell

  • Contributors WvdB, TH, JJvdH and HJCdV conceived this article. RCAA wrote the first draft of the article, with further contributions from all authors. RCAA did statistical analyses with support of JJvdH. All authors interpreted data, reviewed successive drafts and approved the final version of the article.

  • Competing interests None declared.

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