Chlamydia trachomatis (CT) is frequently detected in rectal specimens from asymptomatic sexually-active women who deny receptive anal intercourse (RAI).1–9 Although the natural history of rectal CT infection is unclear, it is thought that the presence of CT in rectal specimens usually reflects shedding of infectious CT organisms from an established rectal focus. Consistent with this interpretation, intrarectal inoculation of mice and non-human primates with Chlamydia establishes long-lasting asymptomatic infection and concomitant shedding of infectious chlamydiae from the rectum.10
Studies primarily of men who have sex with men (MSM) have determined that RAI is an important risk factor for rectal CT, and Neisseria gonorrhoeae (NG) and Mycoplasma genitalium (MG) infection. However, RAI does not explain rectal CT prevalence in women, since rates are similar in high-risk women who do and who do not report RAI.3,7,8,11 Proximity of the urogenital tract and anorectum causes an increased risk for urinary tract infections in women compared with men.12,13 Thus, autoinoculation of the rectum with infected vaginal secretions could also be a source of rectal CT in women who deny RAI. However, a significant proportion of women with rectal CT do not have concurrent urogenital CT,11 which suggests that autoinoculation alone may not fully explain rectal CT rates in women who deny RAI.
An alternate explanation for the detection of rectal CT in individuals who deny RAI has been termed the “oral hypothesis.”10,14,15 This hypothesis proposes that oral sex (cunnilingus, fellatio, or anilingus) efficiently inoculates the gastrointestinal tract with CT and that rectal CT detection may indicate shed CT organisms from 1 or more established gastrointestinal foci. The oral hypothesis is consistent with the known biology of other closely related Chlamydia species for which the fecal-oral route is a primary mode of transmission and the gastrointestinal tract is the site of productive infection.10 In humans, it is unknown if oral inoculation can establish CT gastrointestinal infection or rectal CT shedding. Although most women at risk for CT infection report fellatio,16 the potential contributions of oral exposure and autoinoculation cannot be easily differentiated in women.
Heterosexual men might be a better model for evaluating the oral hypothesis because men commonly engage in cunnilingus, but not RAI or other sexual practices that could directly inoculate the rectum. Rectal autoinoculation by urethral CT also seems less likely in men than in women due to the increased distance between the rectum and urethra. Only a few studies of rectal CT have included heterosexual men. One study of 169 heterosexual men failed to detect any case of oropharyngeal or rectal CT, but only 11 of these men had urogenital CT.4 A retrospective study detected CT in rectal swabs from 3 heterosexual men, but these men reported rectal exposures so direct inoculation could not be excluded.17 A prospective study detected rectal CT in 1.3% (n = 4) of heterosexual male swingers compared with 8.9% of MSM.1 However, specific sexual behaviors were not defined in this study. Another study of male swingers reported that oral and vaginal sex was very common (both >95%) and 61% engaged in anal sex.18 Thus, prior studies of rectal CT in men have not collected sufficient behavioral data to identify all of the potential routes of CT inoculation, especially anal behaviors that could result in direct CT inoculation (anal intercourse, anilingus, and use of sex toys/fingers). Also, reported sexual orientation alone is insufficient to exclude anal behaviors as prior studies have shown that reported sexual orientation may not accurately reflect sexual behaviors.19,20
To address limitations of prior studies of rectal CT in men and evaluate the oral hypothesis, we enrolled men with and without nongonococcal urethritis (NGU) presenting to a sexually transmitted diseases (STD) clinic in Indianapolis, Indiana, and obtained urine and rectal swabs for CT testing. The men completed detailed surveys reporting their sexual orientation and recent and lifetime sexual behavior practices. As an additional control for reported anal exposures, we also tested rectal swabs for NG and MG.
MATERIALS AND METHODS
Study Population and Procedures
As part of an ongoing 5-year study of NGU, the Idiopathic Urethritis Men's Project (IUMP), we recruited males ≥18 years of age with and without urethritis who presented to the Marion County Public Health Department (MCPHD) Bell Flower STD Clinic in Indianapolis, Indiana. Cases were men with NGU, defined as clinician-observed signs and or patient-reported symptoms of urethritis and 5 or more polymorphonuclear cells (PMN) per high-power field by urethral Gram stain smear. Controls were asymptomatic men with no signs or symptoms of urethritis. Men whose urethral smears showed Gram negative diplococci were presumed to have NG infection and were excluded from the study. Interested participants provided written consent and were then enrolled. Men were interviewed and a physical examination performed. Each participant also completed an extensive self-administered computer-assisted survey, which included detailed questions regarding demographics, prior STD history, sexual orientation, and specific recent (ie, within the last year) and lifetime (ie, ever engaged in) sexual behaviors. A first-catch urine and rectal swab were obtained for CT, NG, and MG testing. Men diagnosed with NGU were treated with azithromycin (1 g orally directly observed), advised to refer all sexual partners for treatment, and abstain from sexual activity for 1 week. The study was approved by the Indiana University/Purdue University-Indianapolis Institutional Review Board and by MCPHD.
Pathogen Testing and Calculation of CT Loads
Pathogen-specific nucleic acid amplification and CT load testing was performed by the Indiana University School of Medicine Infectious Diseases Laboratory. The CT and NG testing was performed as previously reported.21 The MG testing was performed using an in-house quantitative polymerase chain reaction assay, as previously described.22 Rectal swab CT load quantitation was derived by comparison to a serial dilution of a known concentration of CT elementary bodies.
All analyses were performed using SAS version 9.4 statistical software. For demographic and medical characteristics, descriptive statistics (medians, ranges, frequencies, and percentages) were computed to assess variables' distributional properties and to describe the sample. Comparisons of the baseline demographics and sexually transmitted infection test results among groups defined by sexual behaviors, self-reported sexual identity, and acceptance of the rectal swab were performed using the Kruskal-Wallis tests for continuous measures, while χ2 tests, or Fisher exact tests if the expected value of cells was below 5, were used for categorical variables.
One hundred ninety-seven men who enrolled in the IUMP study provided a rectal swab and are included in this analysis (Fig. 1). The majority of the men who declined to provide a rectal swab self-identified as heterosexual (n = 108/243, 44.4%). The majority of MSM (n = 62, 96.9%) agreed to provide a rectal swab, with only 3.1% declining. The demographics and clinical characteristics of the self-identified heterosexual men who provided a rectal swab and declined to provide a rectal swab were similar. The reported anal exposures differed in the 2 groups. The men who provided rectal swabs reported receiving anilingus and receptive anal play more frequently (Table S1, https://links.lww.com/OLQ/A372). Characteristics of the study participants are described in Table 1. The median age was 28 years (range, 18–68 years), 46.7% were African American, and 48.2% complained of symptoms of urethritis. One hundred thirty-five (68.5%) identified as heterosexual, 17.8% identified as homosexual, and 13.7% identified as pansexual, bisexual, or asexual. Consistent with the IUMP study design, all the men diagnosed with NGU reported urethral symptoms (Table 1). Abdominal and rectal symptoms were reported in 1% (n = 2) and 1.5% (n = 3) of men, respectively.
Reported Sexual Behaviors in Heterosexual Men
Cunnilingus was the second most frequent lifetime behavior (after vaginal intercourse) in the self-identified heterosexual men (n = 135) with 82% reporting this behavior in the last year and 93.2% reporting lifetime behavior (Table 2). A significant proportion of these men also reported performing anilingus on a female partner in the last year (23.1%) or lifetime (38.1%). In the self-identified heterosexual men, anilingus on a female partner was only reported by men who also reported cunnilingus with female partners. Two (1.5%) of the self-identified heterosexual men reported lifetime RAI. A larger proportion reported non-RAI lifetime receptive anal behaviors with female partners including anal play (20.9%) or receiving anilingus (23.1%) that could potentially inoculate the rectum. Overall, 84 (62.2%) of 135 of the self-identified heterosexual men reported both lifetime cunnilingus and no receptive anal behaviors.
Prevalence of Urethral CT and Rectal CT, NG, and MG
Overall, 37 (18.8%) men had urethral CT, and 21 (10.7%) men had urethral MG (Table 3). No men had urethral NG, consistent with the IUMP study design.
Rectal CT was detected in 8 (4.1%) of 197 men and none of these men reported anorectal symptoms. Six of the anorectal CT cases were in men who were behaviorally bisexual or homosexual. However, among the group of 84 self-identified heterosexual men who reported cunnilingus but no receptive anal exposures in their lifetimes, 2 (2.4%) had rectal CT. One of these 2 men reported cunnilingus within the prior 60 days, whereas the other reported cunnilingus and anilingus with a woman more than a year previously.
Rectal MG was detected in 3 (1.5%) of 197 men. All of these men were behaviorally bisexual or homosexual, with all reporting lifetime anal exposures (included RAI), and the majority reporting these behaviors within the last year. Although men with urethral NG were excluded, consistent with the IUMP design, 3 (1.5%) of 197 men had rectal NG. Similar to the men with rectal MG, all these men reported receptive anal behaviors within the last year. When limited to MSM, prevalence of rectal NG was 4.8% (3/62) and MG was 4.8% (3/62).
In 2 men, CT was detected in both the urethra and rectum. One of these men self-identified as bisexual and reported receptive anal behaviors. The other was 1 of the 2 heterosexual men who denied any receptive anal behaviors.
Calculation of CT Loads
To minimize the possibility that rectal CT results were due to false-positives, we repeated CT testing of the CT-positive rectal swabs and quantified the CT loads (Fig. 2). All of the positive rectal CT tests were confirmed CT-positive by re-testing. In the heterosexual men, one of the CT loads was higher and the other lower than the median of the 6 CT positive swabs from the self-identified homosexual and bisexual men. The confidence intervals were similar between the 2 groups.
In this study, we report detecting rectal CT in 2/84 (2.4%) behaviorally heterosexual men who denied ever engaging in receptive anal behaviors which could directly inoculate the rectum with CT, but who engaged in cunnilingus. This suggests that oral CT inoculation may have occurred in these men, given that the only mucosal sites of exposure were the urethra and oropharynx. In our study, all of the rectal CT cases were asymptomatic, which is consistent a prior study, which reported that asymptomatic rectal CT is common.23
To our knowledge, our study is the first that supports the hypothesis that fecal-oral transmission of CT occurs in humans. Two other lines of evidence suggest that CT can transit the stomach and infect the gastrointestinal tract or rectum in humans. An in vitro study demonstrated that CT can survive incubation at low pH, comparable to gastric environments.24 More importantly, CT DNA has been detected in the appendix and colon25 and CT inclusions were also visualized in the appendix and rectal tissues, confirming that CT can infect these tissues. It is unclear if transit of CT to the appendix and rectal tissues occurred through the fecal-oral route or ascended proximally from rectal inoculation.
Prior studies of rectal CT in heterosexual men had limitations, but support the findings of our study. A review of 5 studies of the prevalence of rectal CT infections in heterosexual men found the median rate was 0.9% (range, 0.0–9.1%), which is comparable with our results.23 In individuals who deny receptive anal sex, the rectal CT prevalence appears to be strikingly lower in men (approximately 1.5% in our study) than in women (13.4%).2 This could reflect a more important role for autoinoculation in acquisition of rectal CT in women compared with men. Indeed, the rate of concordant urogenital and rectal CT infection has been reported to be much higher in women (73%),2 compared with the 3.7% concordance rate in the self-identified heterosexual men in our study (1/27). Another possibility is that the different prevalence of rectal CT in men and women reflects the different risks of oral inoculation posed by fellatio and cunnilingus. Fellatio exposes the oral mucosa directly to epithelial cells of the urethral meatus and to ejaculate. In contrast, cunnilingus primarily exposes the oropharynx to the external genitalia and superficial vaginal surfaces which are not sites of CT infection. Therefore, we posit that the number of infectious CT elementary bodies ingested by individuals performing cunnilingus may be relatively low compared with women who perform fellatio.
We cannot exclude that some heterosexual men in our study inaccurately reported their sexual behaviors, but we attempted to minimize this using multiple approaches. First, sexual behaviors were not an exclusion criterion (ie, no secondary gain was afforded by denying anal sex), and data were collected by computer-assisted self-reporting to minimize bias from sexual behaviors stigma. Nonetheless, due to desirability bias, some of heterosexual men may not have been comfortable reporting receptive anal behaviors even in the setting of an anonymous interview. Second, we compared the demographic and clinical characteristics of the self-identified heterosexual men who provided and declined to provide rectal swabs and found no significant differences between the groups. However, selection bias is possible given that anal exposures were more commonly reported from men who provided rectal swabs. Third, we also tested all men for rectal NG and MG, which have not been documented to establish rectal infection via the fecal-oral route. NG and MG were each detected in 3 homosexual and bisexual men (4.8%) who reported RAI. It is important to note that no men who denied lifetime anal sex, including the 2 men who had rectal CT, had rectal NG or MG detected, which supports their reported behaviors.
In our study, we also quantified the rectal swab CT load to test if our findings could be explained by low level nucleic acid contamination or false-positive tests. It is interesting that compared with rectal CT loads from men who engaged in RAI, and likely had direct rectal inoculation of CT, the median CT load was slightly higher in the 2 men without RAI. Further, the false positive rate of CT nucleic amplification tests is less than 1%26 and all of the rectal CT-positive tests were confirmed to be true-positives by reextracting the original specimens and repeat CT testing.
Our study has strengths and limitations. A major strength is our use of a detailed sexual behavior survey to assess anal exposure behaviors (including anilingus, toys, fingers, etc.) in heterosexual men. Also, we asked about lifetime, in addition to recent, sexual behaviors to minimize confounding by prolonged (>1 year) rectal CT infections or inaccurate reporting of recent anal exposures. Another strength is that our study was controlled by comparison to homosexual men and other pathogens that are introduced into the rectum by RAI (NG and MG). Limitations of our study are that this is a single-site study of men presenting to an STD clinic for evaluation and our findings may not be generalizable to other populations. Also, we used a nucleic acid amplification test to detect rectal CT infections, which cannot differentiate established infections from contamination with non-viable organisms (eg, residual undigested CT DNA passing through the gastrointestinal tract). Our study design excluded men with NGU who had Gram stain PMN counts less than 5 (eg, 2–4 PMNs per high-powered field), which may be a limitation.
In summary, our results support the hypothesis that rectal CT infections can be acquired by oral sex, but suggest that this is relatively uncommon in heterosexual men. Further research is needed to assess the oral route for gastrointestinal infection in both women and men and to define if the infectious inoculum differs between different oral behaviors.
1. van Liere GA, Hoebe CJ, Niekamp AM, et al. Standard symptom- and sexual history-based testing misses anorectal Chlamydia trachomatis
and Neisseria gonorrhoeae
infections in swingers and men who have sex with men. Sex Transm Dis 2013; 40:285–289.
2. van Liere GAFS, Dukers-Muijrers NHTM, Levels L, et al. High proportion of anorectal Chlamydia trachomatis
and Neisseria gonorrhoeae
after routine universal urogenital and anorectal screening in women visiting the sexually transmitted infection clinic. Clin Infect Dis 2017; 64:1705–1710.
3. Gratrix J, Singh AE, Bergman J, et al. Evidence for increased chlamydia case finding after the introduction of rectal screening among women attending 2 Canadian sexually transmitted infection clinics. Clin Infect Dis 2015; 60:398–404.
4. Ostergaard L, Agner T, Krarup E, et al. PCR for detection of Chlamydia trachomatis
in endocervical, urethral, rectal, and pharyngeal swab samples obtained from patients attending an STD clinic. Genitourin Med 1997; 73:493–497.
5. Jenkins WD, Nessa LL, Clark T. Cross-sectional study of pharyngeal and genital chlamydia and gonorrhoea infections in emergency department patients. Sex Transm Infect 2014; 90:246–249.
6. van Liere GA, Hoebe CJ, Wolffs PF, et al. High co-occurrence of anorectal chlamydia with urogenital chlamydia in women visiting an STI clinic revealed by routine universal testing in an observational study; a recommendation towards a better anorectal chlamydia control in women. BMC Infect Dis 2014; 14:274.
7. Musil K, Currie M, Sherley M, et al. Rectal chlamydia infection in women at high risk of chlamydia attending Canberra Sexual Health Centre. Int J STD AIDS 2016; 27:526–530.
8. Ding A, Challenor R. Rectal chlamydia in heterosexual women: More questions than answers. Int J STD AIDS 2014; 25:587–592.
9. Andersson N, Boman J, Nylander E. Rectal chlamydia—Should screening be recommended in women? Int J STD AIDS 2017; 28:476–479.
10. Rank RG, Yeruva L. Hidden in plain sight: Chlamydial gastrointestinal infection and its relevance to persistence in human genital infection. Infect Immun 2014; 82:1362–1371.
11. Chandra NL, Broad C, Folkard K, et al. Detection of Chlamydia trachomatis
in rectal specimens in women and its association with anal intercourse: A systematic review and meta-analysis. Sex Transm Infect 2018; 94:320–326.
12. Griebling TL. Urologic diseases in America project: Trends in resource use for urinary tract infections in men. J Urol 2005; 173:1288–1294.
13. Hooton TM, Scholes D, Hughes JP, et al. A prospective study of risk factors for symptomatic urinary tract infection in young women. N Engl J Med 1996; 335:468–474.
14. Bavoil PM, Marques PX, Brotman R, et al. Does active oral sex contribute to female infertility? J Infect Dis 2017; 216:932–935.
15. Rank RG, Yeruva L. An alternative scenario to explain rectal positivity in chlamydia-infected individuals. Clin Infect Dis 2015; 60:1585–1586.
16. Leichliter JS, Chandra A, Liddon N, et al. Prevalence and correlates of heterosexual anal and oral sex in adolescents and adults in the United States. J Infect Dis 2007; 196:1852–1859.
17. Trebach JD, Chaulk CP, Page KR, et al. Neisseria gonorrhoeae
and Chlamydia trachomatis
among women reporting extragenital exposures. Sex Transm Dis 2015; 42:233–239.
18. Platteau T, van Lankveld J, Ooms L, et al. Sexual behavior and sexually transmitted infections among swingers: Results from an online survey in Belgium. J Sex Marital Ther 2017; 43:709–719.
19. Fu TC, Herbenick D, Dodge B, et al. Relationships among sexual identity, sexual attraction, and sexual behavior: Results from a nationally representative probability sample of adults in the United States. Arch Sex Behav 2018.
20. Dodge B, Herbenick D, Fu TC, et al. Sexual Behaviors of U.S. men by self-identified sexual orientation: Results from the 2012 National Survey of Sexual Health and Behavior. J Sex Med 2016; 13:637–649.
21. Van Der Pol B, Liesenfeld O, Williams JA, et al. Performance of the cobas CT/NG test compared to the Aptima AC2 and viper CTQ/GCQ assays for detection of Chlamydia trachomatis
and Neisseria gonorrhoeae
. J Clin Microbiol 2012; 50:2244–2249.
22. Dutro SM, Hebb JK, Garin CA, et al. Development and performance of a microwell-plate-based polymerase chain reaction assay for Mycoplasma genitalium
. Sex Transm Dis 2003; 30:756–763.
23. Chan PA, Robinette A, Montgomery M, et al. Extragenital infections caused by Chlamydia trachomatis
and Neisseria gonorrhoeae
: A review of the literature. Infect Dis Obstet Gynecol 2016; 2016:5758387.
24. Moncada J, Schachter J. In vitro survival of Chlamydia trachomatis
in a low pH environment. In: Chernesky MA, Clarke IN, Gaydos C, et al, eds. Chlamydial Infections: Proceedings of the fourteenth International Symposium on Human Chlamydial Infections 2018:131–133.
25. Borel N, Marti H, Pospischil A, et al. Chlamydiae in human intestinal biopsy samples. Pathog Dis 2018; 76.
26. Papp JR, Schachter J, Gaydos C, et al. Recommendations for the laboratory-based detection of Chlamydia trachomatis
and Neisseria gonorrhoeae
—2014. MMWR 2014; 63(RR-02):1–19.