Sexually Transmitted Enteric Infections in Men who have Sex with Men

SUMMARY

Sexual transmission of enteric pathogens (STEI) among men who have sex with men (MSM) has been reported since the 1960s and is increasingly recognized since the widespread adoption of multiplex molecular diagnostics. However, the overall burden of sexually transmitted enteric infections has been difficult to ascertain, as the public health response to these infections and identification of transmission networks fall between the traditional groupings of sexually transmitted and foodborne diseases. The global emergence of extensively drug-resistant Shigella and Campylobacter infections among MSM and potential for cross-over between different at-risk populations underscore the importance of timely diagnosis, appropriate treatment, and the need to consider community-level education and testing. Moreover, the possible impact of pre- and post-exposure prophylaxis for HIV and sexually transmitted infections on STEI is presently uncertain. This review examines our evolving understanding of STEI, discusses specific pathogens of urgent importance, and prioritizes areas for further study.

1. Introduction

Men who have sex with men (MSM) are at increased risk for sexually transmitted enteric infections (STEI) (1). The potential for shigellosis and amebiasis to be spread through sexual activity has been recognized since at least 1968 (2). STEI outbreaks have been documented almost exclusively among MSM. Although mechanisms of transmission are incompletely understood, STEI appear to result most often from oral-anal rather than penile contact. While STEI are well recognized, the public health response to these infections is often poorly coordinated, falling between the traditional groupings of sexually transmitted and foodborne diseases. Siloed approaches have left fundamental questions unanswered, including emergent issues relating to molecular diagnostics, control of antimicrobial resistance, and the possible impact of pre- and post-exposure prophylaxis for HIV and sexually transmitted infections on STEI. This review examines our evolving understanding of STEI, discusses specific pathogens of urgent importance, and prioritizes areas for further study.

STEI have been recognized for decades as a global phenomenon. Several case series were reported in New York, San Francisco, London, and Minneapolis in the late 1960s and early 1970s. The earliest report described four MSM who were diagnosed with amebiasis and reported similar sexual contact in New York City (2). In 1974, the first outbreak of enteric infection among MSM was reported in San Francisco. This case series, reported by Selma Dritz and colleagues, described more than 50 people infected with S. flexneri 2a, many of whom were young MSM (3). Considerable efforts to identify a common food-related exposure were unsuccessful. The public health response to the outbreak involved enlisting community organizations, employing direct marketing in targeted neighborhoods, and alerting physicians caring for MSM patients. These multi-pronged efforts were ultimately successful in containing the outbreak.

The impact of sexual transmission on overall enteric infection rates began to gain broad recognition in the 1970s. In 1977, Dritz and colleagues at the San Francisco Department of Health reported a 4- to 10-fold rise in the incidence of shigellosis, amebiasis, and hepatitis A and B infections that was attributable almost entirely to increases among men aged 18-35 (4). The constellation of enteritis, proctocolitis and proctitis in MSM was initially referred to as the “gay bowel syndrome” (5–16), a stigmatizing term that has subsequently fallen out of favor (17, 18) as more has been learned about the specific etiologies of gastrointestinal disease in this population.

Many early studies focused on alerting clinicians to the phenomenon of sexually transmitted enteric pathogens and called for appropriate education regarding abstinence from sex until infections cleared, as well as ensuring comprehensive testing for other sexually transmitted infections (STI). In 1979, Kean, et al. assessed a cohort of men presenting to a community based STI clinic in New York City and found that 39.7% of those evaluated had amebiasis or giardiasis (19). The authors combined these data with known estimates of rates of missed infection on stool examination and rougher estimates of the number of MSM in New York to suggest that between 10,000 and 50,000 New Yorkers might potentially be infected with sexually transmitted amebiasis or giardiasis. Citing “official neglect of the epidemic,” they called on the New York City Health Department to take action and urged a nationwide response to this “newest of old diseases.” The New York City Bureau of Venereal Disease Control took issue with this interpretation and extrapolation of data, as well as the accusation that little was being done in response. Felman and Marr rebutted Kean’s assertions by referencing documented increases in the sexual transmission of enteric pathogens and efforts made by the Health Department since 1977 to address these increases (20). They further asserted that Kean’s cohort was a highly biased sample interpreted using a “histrionic heuristic.” Importantly, this dispute between health authorities and scientists occurred prior to the recognition of HIV/AIDS and presaged the contentious relationship between activists and public health officials that emerged in the early days of the epidemic.

The HIV/AIDS epidemic represented a cataclysmic event that affected nearly all aspects of infectious diseases care for MSM. However, with the notable exception of protozoa (e.g., Cryptosporidium and Cystoisospora) and microsporidia, little difference has been found in the rate of enteric pathogens recovered from stool of HIV-positive and -negative individuals. A 1999 study by Weber, et al. in the Swiss HIV Cohort illustrated that the relative risk of diarrhea in MSM was 2.5 times higher than in men who have sex with women (MSW), even after controlling for CD4 count (21). This study suggested that behavior, not immune status, is the major driver of the increased rate of acute enteritis in MSM-- an observation that remains true in the modern era of multiplex PCR testing and widespread viral suppression with antiretroviral therapy (ART) (22).

Throughout the 1990s, work on the sexual transmission of enteric pathogens largely consisted of cohort studies evaluating the correlation between HIV status, CD4 count, number of sexual partners, age, etc. on the rates of enteric pathogen identification (23–29). In general, higher numbers of sexual partners and young age were associated with higher rates of enteric pathogen detection (30). Intestinal protozoal infection was associated with low CD4 counts (23, 31).

At the turn of the 21^st century, the emphasis in the literature shifted to outbreak reports, primarily involving Campylobacter and Shigella, as well as understanding seroprevalence and rates of Entamoeba histolytica (32–39). These reports highlighted inexorable increases in antimicrobial resistance and the impact of new, more sensitive laboratory diagnostics. Between the 1990s and 2017, the percentage of Campylobacter isolates exhibiting decreased susceptibility to ciprofloxacin nearly doubled, rising from approximately 15% to nearly 30% of isolates tested, with 2% also harboring resistance to azithromycin (40). Nontyphoidal Salmonella also developed increasing resistance to ceftriaxone, ciprofloxacin, and azithromycin; 16% of isolates in 2017 harbored resistance to at least one antibiotic (40). Drug-resistant Shigella infections rose exponentially, with the percentage of cases with resistance to ciprofloxacin or azithromycin increasing by 5- to 10-fold between 2009 and 2017. By 2017, approximately 3% of Shigella isolates were resistant to both ciprofloxacin and azithromycin (40). The importance of international travel in disseminating multidrug-resistant enteric pathogens in MSM sexual networks has been demonstrated by epidemiologic and genomic data (41–43).

In 2022, a Campylobacter infection resistant to all tested antibiotics was detected in an MSM who was coinfected with Mpox (44). The Campylobacter infection cleared spontaneously but raised concerns about the potential for sexual transmission of XDR (extensively drug-resistant) enteric bacteria among MSM. A 2022 retrospective study of MSM with HIV in Italy reported that 75% of Campylobacter isolates were resistant to quinolones, 50% were resistant to tetracyclines, and 17% were resistant to macrolides (45).

Since 2016, XDR Shigella infections, defined as resistant to azithromycin, ciprofloxacin, ceftriaxone, trimethoprim-sulfamethoxazole (TMP-sulfa), and ampicillin, began to be detected in the United States (46). As of 2022, these strains were increasingly common and represented 5% of all isolates tested. XDR Shigella infections were noted to be more prevalent among MSM, persons experiencing homelessness (PEH), international travelers, immunocompromised people, and people with HIV (PWH). XDR Shigella isolates have also been detected in MSM in Australia, England, and across continental Europe (41, 47–49).

2. Diagnosis of Enteric Infections in MSM

Acute gastroenteritis produces similar signs and symptoms regardless of etiology, but timely identification of infectious causes of gastroenteritis can influence patient counseling and options for treatment. Historically, traditional laboratory diagnostic methods to identify the causes of acute infectious diarrhea have consisted of a combination of culture on selective media, antigen testing, and microscopy. However, such methods are laborious and require a turnaround time of days for final results, limiting their utility for guiding initial treatment decisions. Although biomarkers of host inflammation such as fecal calprotectin have some ability to discriminate between individuals with or without enteric infection, they are unable to distinguish specific bacterial pathogens or between bacterial and viral infections (50), which may warrant different antimicrobial therapies. Thus, clinical laboratories are increasingly using nucleic acid amplification tests (NAATs) to identify specific pathogens rapidly and accurately in fecal samples. Multiplex PCR panels can simultaneously detect multiple pathogens and presently range in size from four to more than twenty microbial targets, with results available within hours. These multiplex tests do not detect N. gonorrhoeae, Chlamydia trachomatis, Treponema pallidum or herpes simplex virus, which must be tested for separately.

Since the xTAG GPP (Luminex) became the first FDA-cleared multiplex test for the diagnosis of enteric pathogens in 2013 (51), additional commercial multiplex PCR panels have become available (52). Studies show that in addition to dramatically reducing turnaround time, multiplex panels have greatly improved the sensitivity of diagnostic testing in patients with acute infectious gastroenteritis (53). Moreover, multiplex panels may enhance both diagnostic and antimicrobial stewardship. For instance, the rapid detection of viral pathogens can obviate the need for empirical antimicrobial therapy. In clinical practice, large multiplex panels (containing ≥12 targets) have specifically been associated with reduced antibiotic prescribing, fewer follow-up visits, and less diagnostic imaging, without increasing overall health care costs (54). It is not unusual for multiplex panels to detect more than one enteric pathogen, particularly in MSM, travelers recently returning from low- or middle-income countries (LMICs), and PEH (22, 53) [Table 1]. For certain reportable bacteria, such as Campylobacter spp., Shiga toxin-producing E. coli, Salmonella enterica, and Shigella spp., reflexive culturing should be performed whenever possible to allow strain typing and/or determination of antimicrobial resistance (55).

Table 1. Prevalence of enteric pathogens among men who have sex with men (MSM) in studies utilizing multiplex PCR.

Total represents number of samples included in the study, which was ≥1 per person for some studies. Positive represents number of samples testing positive for ≥1 pathogen.

	Symptomatic Individuals					Asymptomatic Individuals
Ref.	Newman 2020	Purpura 2021	Chow 2021	Verma 2022	Miller 2024	Williamson 2019	Tsai 2024
Location	Seattle, WA	New York, NY	Melbourne, Australia	New York, NY	Seattle, WA	Melbourne, Australia	Taiwan
Population	Symptomatic MSM	MSM on PrEP	Symptomatic MSM with proctitis	Symptomatic MSM with HIV	Symptomatic MSM	Asymptomatic MSM	Asymptomatic MSM
Test(s)	BioFire GI Panel+/-culture	BioFire GI Panel	AusDiagnostics Faecal Pathogen M 16-well assay	BioFire GI Panel	BioFire GI Panel	AusDiagnostics Faecal Pathogen M 16-well assay	BD Max™ Enteric Bacterial Panel and BD Max™ Enteric Parasite Panel
Pathogen	n=268 total	n=194 total	n=499 total	n=126 total	n=436 total	n=519 total	n=375 total
	n=151 positive	n=62 positive	not reported	n=79 positive	n=271 positive	n=57 positive	n=27 positive
Bacteria
Aeromonas spp.	-	-	0	-	-	2 (0.4)	-
Campylobacter	26 (9.7)	14 (7.2)	12 (2.4)	15 (11.9)	17 (3.9)	13 (2.5)	3 (0.8)
Clostridium difficile	9 (6.0)	0	1 (0.2)	2 (1.6)	15 (3.4)	0	-
Escherichia coli
EAEC	29 (10.8)	16 (8.2)	-	14 (11.1)	62 (14.2)	-	-
EPEC	29 (10.8)	20 (10.3)	-	19 (15.1)	78 (17.9)	-	-
ETEC	5 (1.9)	1 (0.5)	-	3 (2.4)	14 (3.2)	-	-
STEC	1 (0.3)	0	12 (2.4)	1 (0.8)	0	9 (1.7)	2 (0.5)
Shigella	46 (17.2)	4 (2.1)	14 (2.8)	19 (15.1)	76 (17.4)	5 (1.0)	5 (1.3)
Salmonella	1 (0.3)	0	1 (0.2)	0	2 (0.4)	2 (0.4)	3 (0.8)
Vibrio	0	2 (1.0)	0	0	1 (0.2)	0	-
Yersinia	2 (0.7)	0	0	1 (0.8)	1 (0.2)	9 (1.7)	-
Viruses
Adenovirus	1 (0.3)	0	0	1 (0.8)	1 (0.2)	0	-
Astrovirus	2 (0.7)	0	0	2 (1.6)	3 (0.7)	10 (1.9)	-
Norovirus	23 (8.6)	2 (1.0)	3 (0.6)	21 (16.7)	51 (11.7)	2 (0.4)	-
Rotavirus	5 (1.9)	0	0	1 (0.8)	8 (1.8)	3 (0.6)	-
Sapovirus	8 (3.0)	1 (0.5)	2 (0.2)	1 (0.8)	10 (2.3)	0	-
Protozoa							-
Blastocystis hominis	6 (2.2)	-	0	-	-	-	-
Cryptosporidium	9 (3.4)	2 (1.0)	1 (0.2)	11 (8.7)	17 (3.9)	0	0
Dientamoeba fragilis	1 (0.3)	-	0	-	-	0	-
Entamoeba histolytica	1 (0.3)	0	0	0 (0)	1 (0.2)	2 (0.4)	6 (1.6)
Giardia	31 (11.6)	19 (9.8)	5 (1.0)	15 (11.9)	42 (9.6)	7 (1.3)	8 (2.1)

Abbreviations: PrEP=pre-exposure prophylaxis for HIV; EAEC=enteroaggregative E. coli; EPEC=enteropathogenic E. coli; ETEC=enterotoxigenic E. coli; STEC=Shiga-like toxin-producing E. coli.

Interestingly, as diagnostic specificity for enteric pathogens has improved over time and more granular clinico-epidemiologic data have accumulated, reports in the literature have recapitulated evidence from older studies that described increased rates of enteric pathogens among MSM and confirmed sexual transmission as a significant mode of transmission. Concomitant or prior STI are common in MSM with sexually transmitted enteric infections (44, 56–58). Two contemporary reports from the United Kingdom published in 2018 (59, 60) echo the findings of Dritz and colleagues from the 1970s (4). The relative contribution of sexual transmission of enteric pathogens compared to other modes of transmission can be obtained by evaluating the number of excess infections among men. The disproportionate number of enteric infections in MSM cannot be accounted for by outbreaks alone and suggests that ongoing endemic transmission is most likely contributing to the excess morbidity observed. While sexual transmission of enteric pathogens is not a new phenomenon, we have yet to fully elucidate the role it plays at the population level in relation to foodborne and other modes of transmission. Even in the modern era, lack of information on sexual history remains a significant limitation on data reported in the medical record and in the literature (61, 62).

3. Recognized Sexually Transmitted Pathogens

a. Campylobacter spp.

Campylobacter is a genus of curved Gram-negative bacteria that are found in the intestinal tracts of many mammals and birds. These organisms are microaerophilic and culturable on selective media. Campylobacter jejuni is the species most frequently associated with diarrheal disease, but C. coli and C. upsaliensis are also common causes of enteritis in humans. Early reports described bloodstream infections in MSM caused by “Camplyobacter-like organisms” that have since been re-classified as Helicobacter cinaedi, Helicobacter fennelliae, and CLO-3 (63–67). In the U.S., campylobacteriosis is a nationally notifiable disease.

Classically, Campylobacter infections are associated with diarrhea and abdominal pain, sometimes preceded by a prodromal fever. Among MSM, Campylobacter is sexually transmissible and can produce proctitis, with or without colitis and diarrhea (68). A case series describing clinical manifestations in patients at sexual health clinics has detailed that anorectal pain, discharge, and tenesmus are frequently found (66). Campylobacter infections are often self-limited and resolve with supportive care alone but may be more severe in PWH (69). Important late complications include reactive arthritis, Reiter’s syndrome, hemolytic-uremic syndrome, irritable bowel syndrome, and Guillain-Barré or Miller Fisher syndrome resulting from molecular mimicry of human gangliosides by Campylobacter lipooligosaccharide (70). Macrolides are most often used for treatment, which can decrease the duration of symptoms and bacterial shedding following acute illness (71) [Table 2].

Table 2.

Recommended treatment regimens for select sexually transmitted enteric pathogens.

Pathogen	Treatment Indications	Treatment	AMR Considerations	Possible Alternative Agents
BACTERIA
Campylobacter	Severe or prolonged symptoms, immunocompromised host; consider treatment to prevent transmission	Macrolides preferred; fluoroquinolones are alternative	Fluoroquinolone resistance rising; XDR strains reported	Carbapenems, fosfomycin
E. coli	Avoid antibiotics for STEC; Consider antibiotics for severe or prolonged symptoms, immunocompromised host caused by EAEC/EPEC	Fluoroquinolones, macrolides; susceptibility testing not available	Resistance rising globally	Rifaximin, fosfomycin
Salmonella, nontyphoidal	Avoid antibiotics in uncomplicated infections; Consider antibiotics for severe or prolonged symptoms, immunocompromised host, ≥50 years of age	Fluoroquinolones, macrolides; ceftriaxone for hospitalized patients	Resistance to fluoroquinolones rising	Ertapenem
Salmonella, Enteric fever	All infections	Fluoroquinolones, azithromycin; ceftriaxone for hospitalized patients	Resistance to fluoroquinolones and 3rd generation cephalosporins rising; XDR strains reported	Ertapenem
Shigella	Treat all infections in MSM	Fluoroquinolones, azithromycin, TMP-sulfa depending upon local susceptibility patterns; Ceftriaxone for hospitalized patients	TMP-sulfa resistance common; Resistance to fluoroquinolones, macrolides and 3rd generation cephalosporins rising; XDR strains reported	Fosfomycin, pivmecillinam, ertapenem
PARASITES
Cryptosporidium	Usually supportive treatment; Immune restoration is most effective in people with HIV	Nitazoxanide; not effective in people with low CD4 counts
Cystoisospora	Severe or prolonged symptoms, immunocompromised host	TMP-sulfa; consider secondary prophylaxis
Entamoeba histolytica	Treat all infections in MSM (but not infections with non-pathogenic Entamoeba sp.)	Metronidazole or tinidazole AND paromomycin; Paromomycin alone for asymptomatic cyst carriage
Enterobius	Treat symptomatic infections	Mebendazole, albendazole, pyrantel pamoate
Giardia	Treat all infections	Tinidazole; alternative is metronidazole	Resistance to nitroimidazoles rising	Quinacrine, paromomycin, nitazoxanide, combinations with albendazole
Microsporidiosis	Treat symptomatic infections	Albendazole		fumagillin
Strongyloides	Treat all infections	Ivermectin		albendazole

The first report of sexual transmission of Campylobacter was published in 1979 (72). Since then, it has been consistently observed in prospective cohort and cross-sectional surveys in both symptomatic and asymptomatic MSM. A 1984 cross-sectional study showed increased rates of Campylobacter jejuni IgG positivity in MSM attending an STI clinic when compared to attendees of other clinics (73). Modern studies using NAAT-based diagnostic methods show that up to 17% of MSM presenting with diarrhea are positive for Campylobacter species (22). Outbreaks of C. jejuni have been documented among MSM in Quebec and the United States (36, 74, 75). Identified risk factors for Campylobacter spp. infection in MSM include HIV infection, urban residence, use of HIV-PrEP, concurrent STI, prior STEI, condomless receptive anal sex, oral-anal sex, multiple new sexual partners, and visiting sex-on-premise venues (76). Asymptomatic carriage may be associated with proton-pump inhibitor use (77).

Globally, AMR in Campylobacter spp., particularly to fluoroquinolones, is rising in many settings (36, 44, 74, 75, 78–80). In 2016, an outbreak of a C. coli strain resistant to erythromycin, tetracycline, and ciprofloxacin was reported among 6 MSM in Montreal (81). Subsequently, a study conducted in Montreal and Seattle used whole-genome sequencing to reveal identical strains of MDR (multidrug-resistant) C. coli circulating in both cities over a 4-year period. This strain was also isolated from sporadic cases in the midwestern United States (75). A 2022 retrospective study of MSM with HIV in Italy reported that 75% of Campylobacter isolates were resistant to quinolones, 50% were resistant to tetracyclines, and 17% were resistant to macrolides (45). In 2022, an MSM with Mpox disease was also diagnosed with a Campylobacter infection resistant to all tested antibiotics that spontaneously resolved without intervention (44).

b. Diarrheagenic Escherichia coli

The diverse enteric Gram-negative bacterial species Escherichia coli contains both commensal residents of the intestinal tract and pathotypes that cause disease. E. coli strains that cause enteric infections are collectively referred to as diarrheagenic E. coli (82, 83). The virulence of diarrheagenic E. coli is conferred by the accessory genome— specifically, horizontally-acquired genetic elements in the form of plasmids or genomic islands (84). Enteroinvasive E. coli (EIEC) carries a large virulence plasmid that is also found in Shigella, which allows it to cause inflammatory enterocolitis that is indistinguishable from shigellosis (85). Shiga toxin-producing E. coli (STEC) carries one or two phage-encoded Shiga-like toxins (SLT1 and SLT2) and causes hemorrhagic colitis and hemolytic-uremic syndrome (86). Enterotoxigenic E. coli (ETEC) produces heat-labile (LT) or -stable (ST) enterotoxins, which subvert host cyclic nucleotide pathways to cause secretory diarrhea (87). Enteropathogenic E. coli (EPEC) possesses a LEE (locus of enterocyte effacement) pathogenicity island encoding a type III secretory system that allows the bacterium to interact intimately with host intestinal cells to induce cytoskeletal rearrangements and cause “attaching and effacing lesions” of the gut mucosa (88). Enteroaggregative E. coli (EAEC) carries a large plasmid that allows it to attach to the mucosa and produce enterotoxins and cytotoxins (89). A sixth pathotype called diffusely-adherent E. coli (DAEC) has been described, but its role in diarrheal disease is controversial (90). Hybrid strains also exist, such as the E. coli O157:H7 strain, which carries both Shiga toxins and the LEE pathogenicity island, and the E. coli O104:H4 strain responsible for a large 2011 hemolytic-uremic syndrome outbreak in Germany, which carries both Shiga toxins and the EAEC plasmid (91).

An important role of diarrheagenic E. coli in LMICs (particularly in young children), travelers, and foodborne illness has been clearly demonstrated (92). Although asymptomatic carriage of diarrheagenic E. coli is common, and most infections are self-limited without antimicrobial therapy, EAEC and EPEC are capable of biofilm formation and can cause chronic diarrhea (93–96). The importance of diarrheagenic E. coli in MSM is poorly understood, in part because conventional laboratory diagnostics are unable to distinguish most diarrheagenic E. coli strains from commensal E. coli. However, the increasing use of multiplex PCR panels by clinical laboratories is providing evidence that diarrheagenic E. coli are highly prevalent in MSM and in the general population of high-income countries (22, 53). Molecular markers of STEC, ETEC and EPEC have been identified and used to identify these pathogens on commercial panels (97). EAEC are highly heterogenous and may require multiple markers for definitive identification.

Recent studies suggest that diarrheagenic E. coli are among the most prevalent enteric pathogens in MSM with gastroenteritis (98), as they are in the general population (53, 99). Diarrheagenic E. coli (EPEC and EAEC) were found in 33.1% of samples in MSM who underwent stool testing at University of Washington-affiliated hospitals in 2017-2018, making them the most frequently detected bacterial pathogens in that study (22). A study of 298 persons with HIV presenting with acute diarrhea in New York City, 43.3% of whom were MSM, also found EPEC and EAEC to be among the most common bacteria detected by multiplex panel, comprising 11.1% and 6.0% of samples tested, respectively, with similar rates irrespective of CD4 count (100). MSM (31.7%) were more likely to carry E. coli than non-MSM (19.4%) and also had higher overall rates of infectious diarrhea. EAEC or EPEC are more often detected in non-hospitalized patients (100) and tend to be associated with longer symptomatic intervals prior to clinical presentation (53). STEC is somewhat less prevalent in MSM, but STEC O111:H7 was responsible for an outbreak involving eight MSM in the UK in 2013-2014, possibly imported from South America (101, 102). One patient in that outbreak had an acute onset of bloody diarrhea, while others had milder and more persistent gastrointestinal and constitutional symptoms that lasted for weeks.

Diarrheagenic E. coli colonization is frequently found in asymptomatic individuals (103), and this is true for MSM as well. A convenience sample of rectal swabs from MSM seen at an STI clinic for rectal Chlamydia infection contained EAEC in 5.2%, and 9 of the 17 individuals with EAEC were asymptomatic (104). Surveys of rectal samples detected EPEC or EAEC in 10.3% and 8.2% of asymptomatic MSM on PrEP in New York City, (105), and in 1.7% and 4.9% of MSM attending an STI clinic in London (106). Asymptomatic STEC carriage has also been detected in MSM, albeit less frequently (77, 106–108).

Antimicrobial resistance is widespread in E. coli, and this includes diarrheagenic strains (109–113). Genes encoding extended-spectrum beta-lactamases (ESBL) or other mechanisms of antimicrobial resistance have been found in stool samples from MSM (114–116), and the O111:H7 STEC strain responsible for the UK outbreak was macrolide-resistant (117). Due to a general lack of cultured isolates and molecular epidemiology studies, little is presently known about the diversity and resistance profiles of strains circulating in sexual networks of MSM. Antibiotic treatment is generally not recommended for diarrheagenic E. coli infections (118), although treatment has been beneficial in cancer patients (119) and warrants consideration in individuals with prolonged symptoms. Antibiotics are generally contraindicated in STEC infections because of their potential to induce the hemolytic uremic syndrome (120, 121), although azithromycin was associated with a reduction of STEC carriage during the 2011 German outbreak (122).

Risk factors for diarrheagenic E. coli in MSM include behavior factors as seen for other enteric infections, such as the use of HIV PrEP or dating apps, group sex, high numbers of partners, oral-anal or penile-anal sex, sex toys, STI or other enteric coinfections, and chemsex (the use of recreational drugs to enhance sexual activity) (1, 98). Temperature has been implicated as a driver of diarrheagenic E. coli, with every 1°C in mean monthly temperature correlating with an 8% increase in incidence (123), suggesting a potential impact of climate change.

c. Salmonella enterica

Salmonella enterica comprises a ubiquitous and diverse species of Gram-negative, rod-shaped, facultative anaerobic enteric bacteria subdivided into thousands of serovars (also called serotypes) based on antigenicity. Serovars are also grouped based on host tropism and clinical phenotype into typhoidal and nontyphoidal Salmonella. Studies during the past 15 years indicate that typhoidal Salmonella, which includes the Typhi and Paratyphi serovars, is estimated to cause 15-29 million cases per year globally and up to 200,000 deaths (124). In 2010, nontyphoidal Salmonella was estimated to cause 62-132 million cases per year globally and 155,000 deaths (125), many occurring in sub-Saharan Africa (126). Salmonella has many animal reservoirs, including birds, pigs, cattle, cats, dogs, and reptiles, and most serotypes exhibit a broad host range (127). Transmission to humans occurs through direct contact with animals or animal manure, consumption of contaminated food or water, or from person-to-person contact. Sexual transmission has been documented (128) but is uncommon (45, 129). Salmonella can persist in the environment and has been cultured from groundwater a month after inoculation (130). Salmonellosis is a reportable condition in the United States.

The clinical presentation of salmonellosis varies by serovar. Typhoid (enteric) fever caused by the Typhi or Paratyphi serovars, for which humans are the only host, is a severe illness characterized by fever, headache, and malaise, as well as a range of nonspecific symptoms, including abdominal pain, diarrhea, or constipation (131). After ingestion, bacteria invade through the intestinal epithelium, leading to bacteremia and systemic dissemination, including into the bone marrow (132). Diagnosis is usually made by blood culture (133). The incubation period is 4-14 days, but can be longer, with bacteria often detectable in stool before the onset of symptoms (134). Acute illness lasts 7-10 days following the initiation of antibiotic treatment. Antibiotics do not reliably suppress Salmonella shedding in stool and may prolong it (135) Historically, 2-5% of people with acute typhoid fever go on to become chronic carriers of S. Typhi (136) who shed bacteria in their stool indefinitely. Untreated typhoid fever has a fatality rate of 10-30%, which falls to less than 1% with appropriate treatment (131). Treatment of typhoid fever with antibiotics is recommended to reduce the duration of illness and likelihood of death. Vaccination against S. Typhi is available and recommended for travelers to endemic areas and in low- and middle-income countries where typhoid fever is prevalent.

Nontyphoidal salmonellosis usually presents as a less severe clinical illness. The most common presentation is acute gastroenteritis with diarrhea, abdominal pain, and fever beginning 12-72 hours after exposure. Symptoms typically resolve in 1-7 days without treatment. Invasive nontyphoidal salmonellosis occurs in approximately 5% of cases, with a higher likelihood in people who are immunosuppressed, including those with HIV (137), older adults, and infants. Invasive illness can be systemic with bacteremia and sepsis or focal with osteomyelitis, septic arthritis, meningitis, or other presentations. Aortitis is a serious complication seen in older individuals (138). Diagnosis is usually made by stool culture or PCR, and invasive infection can be identified by culturing blood, tissue, or other bodily fluids. Reflexive culture is recommended if infection is diagnosed by molecular methods to allow susceptibility testing and serotyping. Treatment of non-invasive nontyphoidal salmonellosis is typically supportive. Antibiotics are not recommended in people without risk factors for invasive disease because they do not reduce symptoms and can increase the duration of bacterial shedding; stool culture positivity 1 month post-infection is twice as likely in patients receiving antibiotics (135). For adults at risk of invasive disease or who are suspected to have invasive disease, empiric treatment with a fluroquinolone or ceftriaxone may be used but should be tailored based on culture results due to high rates of antibiotic resistance. For returning travelers from South Asian countries, particularly Iraq and Pakistan where extensive drug resistance is reported in Salmonella Typhi and Paratyphi serovars, empiric treatment of mild illness with azithromycin and severe illness with a carbapenem is recommended. For other travelers, treatment as for non-typhoidal Salmonella is appropriate. Treatment should always be tailored based on culture and susceptibility data when available.

Although there have been well documented instances of sexual transmission of Salmonella, sexual transmission remains relatively uncommon at a population level. Sexual transmission of typhoidal Salmonella was first documented in 1977 in two separate cases of MSM in San Francisco whose partners who were found to be chronic carriers (128). Isolated outbreaks of Salmonella Typhi among MSM with a common partner who was an asymptomatic carrier have been reported as well (37). Subsequent studies of enteric infections among MSM and MSMW have identified additional cases, though often at similar rates to control groups (22, 139, 140).

Population-level data regarding the association between Salmonella infection and MSM have historically been confounded by the impact of HIV/AIDS. During the 1980s, a significant increase in the incidence of Salmonella bacteremia in the U.S. was observed in men aged 25-49 in states with a high incidence of AIDS (141). The increase involved multiple nontyphoidal serovars, and was observed in women as well, making sexual transmission less likely. An additional contemporaneous analysis of the risk of recurrent Salmonella sepsis among people with AIDS found that salmonellosis was more common among adult persons who inject drugs (PWID) or heterosexual HIV acquisition than among MSM or MSMW (141). Data from the Netherlands found a decrease in the reported incidence of nontyphoidal salmonellosis among people with HIV from 1991 to 2003 (142), coinciding with the introduction of ART and the use of trimethoprim-sulfamethoxazole as Pneumocystis prophylaxis (143, 144). There has been some suggestion that typhoidal Salmonella may exhibit more sexual transmission. A study from England found nontyphoidal Salmonella to be equally common among men and women, but typhoidal Salmonella exhibited a male predominance, similar to other known STEI, although this was most pronounced in the 2000s; after 2010, excess male cases were no longer observed (60). Additional contemporary surveillance data from Denmark found that MSM were more likely than controls to be diagnosed with Shigella or Campylobacter but not Salmonella, suggesting that salmonellosis may be significantly less likely to spread through sexual transmission than certain other enteric pathogens (129).

Multiple possible explanations for the lower incidence of sexually transmitted Salmonella infections have been suggested. One reason is that Salmonella has a higher infectious dose than Campylobacter or Shigella (145), which can vary depending on the serovar and vehicle (134, 146). An additional factor may be the lower rate of asymptomatic infection with Salmonella compared to other STEI. A 2019 study of MSM in Australia identified two cases of asymptomatic Salmonella spp. detection out of 512 rectal swab samples (0.4%), whereas five swabs (1.0%) were positive for Shigella spp. and 13 (2.5%) for Campylobacter spp. (107). A 2022 study from the Netherlands failed to identify cases of Salmonella infection among 212 asymptomatic MSM and 177 symptomatic or recently symptomatic MSM, despite multiple cases of Shigella and Campylobacter among both groups (147). These studies and others have been summarized in a recent meta-analysis of asymptomatic STEI infection among MSM (148). Additional potential reasons for the lack of observed sexual transmission on a population scale may be because nontyphoidal Salmonella has a brief duration of illness and shedding in immunocompetent adults, the availability of vaccines for S. Typhi, the use of Pneumocystis prophylaxis in persons with HIV/AIDS, and a lack of data from MSM in areas of high typhoidal Salmonella prevalence, such as Pakistan and India. In addition, the gut microbiome is an important determinant of colonization resistance to Salmonella (149), and it is conceivable that microbiome alterations observed in MSM (150) may be influencing susceptibility to salmonellosis, as has been suggested for C. difficile infections (see below).

d. Shigella spp.

The bacterial genus Shigella includes four species of enteric Gram-negative rods that cause significant illness in humans: S. dysenteriae, S. flexneri, S. boydii and S. sonnei (corresponding to serogroups A-D). Although S. boydii and S. dysenteriae are important causes of shigellosis in LMICs (151, 152), S. flexneri and S. sonnei are the predominant species associated with sexual transmission. More than 40 serotypes of Shigella have been described, but most infections in MSM are caused by the S. sonnei unique serotype and S. flexneri serotypes 2a and 3a (42, 102, 153). Shigella flexneri serotype 3a was responsible for an epidemic in MSM in England from 2009-2014 and subsequently declined in incidence, only to re-emerge from 2019-2021 despite COVID restrictions (154). Most re-emergent isolates were closely related to the original epidemic strain, and no demographic features to account for re-emergence have been identified.

Shigellosis is spread through a fecal-oral route and manifests a spectrum of gastrointestinal illness ranging from asymptomatic infection to self-limiting diarrhea to life-threatening dysentery with fever and bloody diarrhea (155). Although Shigella infections are usually regarded as self-limiting and short-lived, persistent carriage has been observed in MSM (156). Illness can be further complicated by sepsis, bacteremia, hemolytic uremic syndrome, reactive arthritis, and toxic megacolon. In MSM, frequent presentations of shigellosis include anal pain, proctitis, or diarrhea (108, 157, 158) , although more serious illness leading to hospitalization is not uncommon (159). In MSM with advanced HIV (CD4 count <200), shigellosis may be more likely to cause severe illness (160–162). Although the majority of adults do not require hospitalization, adult men are more likely to be hospitalized for shigellosis (157, 163), and data from England suggest that MSM are at higher risk for severe S. flexneri infection (164). Some outbreak strains may result in higher hospitalization rates-- 8 of 33 men (24%) were hospitalized due to a 2021-2022 outbreak of XDR S. sonnei in England (159).

A recent resurgence in shigellosis has been observed in the U.S., Europe and Taiwan, particularly driven by MSM and PEH, accompanied by a rise in antimicrobial resistance (165). In the Netherlands, more than half of domestic shigellosis cases occur in MSM (43). Shigella was one of the first enteric pathogens to be linked to sexual transmission in MSM (3, 166, 167). In high-income countries, MSM represent a significant proportion of reported cases. In the United States, rates of shigellosis among men aged 20-49 rose 5-fold between 1975 and 1985 (168), likely due to spread in the MSM community. From 1989 to 2000, shigellosis rates among adults of both sexes declined, but from 2000-2002 an increase in the proportion of adult male cases was observed contemporaneously with multiple large outbreaks detected in MSM (33, 169). Incidence in men again rose substantially from 2007-2016, coinciding with rises in other bacterial STI (46, 170, 171). U.S. surveillance data from the past nine years have not yet been published. Male predominance in adult cases of shigellosis has also been seen in England, with a significant rise in MSM cases (59, 60, 172–175). In Australia, S. sonnei is more common among men than women, but the gender discrepancy is less pronounced for S. flexneri, suggesting possible network-level effects (176). Studies from the Netherlands between 1992-2006 showed an overall decrease in shigellosis cases but substantial fluctuation over time (177), which may indicate more complex strain dynamics or episodes of strain introduction and loss from a larger pool of strains, rather than consistent endemic circulation of a single strain. In Vancouver, as in Seattle, MDR Shigella strains circulating among MSM have subsequently emerged in PEH, resulting in increasing incidence and severity of infections (178, 179). Although international transmission of Shigella strains among MSM has been documented (174, 180–182), the dynamics of transmission within high-income countries, including between MSM and other demographic groups, likely vary but have been inconsistently reported.

Shigella transmission is facilitated by a low infectious dose, prolonged shedding, asymptomatic infection, and environmental persistence. The median infectious dose (ID50) of Shigella across different strains and species is approximately 1-3x10³ colony forming units (CFU), and ingestion of as few as 10 CFU is sufficient to cause symptomatic infection (183–185). Shigella is shed at mean levels of 10⁵-10⁶ CFU per gram of stool during illness (186) and at lower levels for a month or more after infection (187). Some adults can become long-term carriers following infection, with viable bacteria present in stool for over a year (188). Shigella can also be detected in stool 2-3 days prior to the onset of illness (187). Challenge studies and epidemiologic research in MSM suggest that 2-10% of adult men infected with Shigella may be asymptomatic (147, 186). A recent systematic review and meta-analysis estimated that among asymptomatic MSM visiting STI clinics, the prevalence of Shigella is approximately 1.1% (148). The rate of asymptomatic Shigella carriage may be elevated for MSM who are taking proton pump inhibitors (77).

Risk factors for sexual acquisition of Shigella among MSM include biological, behavioral, and environmental factors. Sexually transmitted shigellosis is associated with HIV infection, having other STI, taking HIV pre-exposure prophylaxis, meeting partners through mobile phone applications or sex on premise venues like gay bathhouses, chemsex and recreational drug use, multiple non-regular sexual partners, oral-anal sexual contact, and living in a large urban area (189) . Among MSM with HIV, lower CD4 count, higher HIV viral load, and not being on antiretroviral therapy are associated with sexually transmitted shigellosis. It is unclear if the increased risk of shigellosis in this group is due to immunosuppression from HIV or indicative of decreased engagement in care and increased behavioral risk factors for Shigella acquisition. Urban areas may increase risk via access to a higher density of MSM and larger network for continued circulation of strains, as seen for other STI.

The diagnosis of Shigella infection is made by stool culture or culture-independent diagnostic tests (CIDT) such as PCR. CIDT are more sensitive than culture and may have contributed to increased case detection in the past decade (176). If a CIDT is used for diagnosis, the CDC (Centers for Disease Control) recommends sending stool for culture to confirm the diagnosis and obtain antibiotic susceptibility testing in view of rising rates of antibiotic resistance (190). Shigella readily acquires resistance to antibiotics. Through sanitation measures and antibiotic use, the incidence of shigellosis declined significantly during the mid-20^th century, but by the 1960s, S. sonnei resistant to ampicillin, sulfonamides, streptomycin, and tetracycline had already been identified in England (191). An epidemic strain of S. dysenteriae with resistance to sulfathiazole, chloramphenicol, tetracycline, and ampicillin emerged in the Western Hemisphere in the late 1960s (192) and spread to Central Africa, where resistance to trimethoprim and nalidixic acid developed after use for treatment (155). Antibiotic-resistant strains of S. flexneri and S. sonnei intensified in the late 20^th century and were strongly associated with sexual networks of MSM. The first report of antibiotic-resistant Shigella transmission among MSM was published in 1977 and detailed an increase in tetracycline-resistant S. flexneri and S. sonnei in Seattle (167).

Since that time, outbreaks of increasingly more resistant Shigella strains have been repeatedly identified among MSM. International transmission of azithromycin-resistant shigellosis among MSM (42) with subsequent acquisition of resistance to ciprofloxacin and beta-lactams has led to XDR strains that are resistant to all conventional anti-Shigella agents (41). In the United States, England, and Western Europe, antibiotic-resistance is more likely to be found in Shigella spp. associated with MSM (193–196). WGS (whole genome sequencing) has demonstrated widespread international circulation of multidrug-resistant clones of Shigella among MSM (43, 178, 197, 198), as well as novel regional lineages (199). Resistance may be acquired during prolonged periods of Shigella carriage, through reinfections, or in the course of prolonged transmission within a sexual network, which in turn increases the likelihood of antimicrobial exposure to agents administered for other STI, acquisition of resistance genes from the commensal gut microbiome or other Shigella strains, and within-host evolution (200, 201) [Figure 1].

Figure 1:

Factors contributing to higher prevalence of multidrug resistant organisms among sexual networks of MSM

In typical cases, treatment of shigellosis is supportive. However, appropriate early antimicrobial treatment can decrease symptom duration and decrease the likelihood of forward transmission by limiting the duration of shedding (202). The latter is of particular importance in MSM, in whom the route of spread is from person-to-person. Shigellosis may be treated with ciprofloxacin, TMP-sulfa, ampicillin, ceftriaxone or azithromycin, depending upon local susceptibility patterns and disease severity (118, 203), but resistance to one or more of these agents is common in MSM. Resistance to FQs and macrolides is now widespread in Shigella in MSM (204), and resistance to 3^rd generation cephalosporins due to blaCTX-M15 or blaCTX-M27 is increasing. Carbapenems, fosfomycin, pivmecillinam and tigecycline are possible alternative agents for XDR infections (205). AMR, treatment failure, and HIV co-infection may contribute to prolonged symptomatic or asymptomatic shedding and transmission among MSM (206). The CDC recommends the use of local resistance susceptibility data in combination with sexual and travel history to guide empiric treatment (207). This may be further complicated by concurrent circulation of S. flexneri and S. sonnei strains with different resistance patterns within the same region.

Due to delays in obtaining antibiotic susceptibility testing results, some authors have proposed the use of local antibiograms and patient-level factors to help identify risk groups that correlate with antimicrobial susceptibility. During an outbreak of shigellosis among MSM and PEH in Seattle (178), experts recognized that isolates in MSM were largely ciprofloxacin-susceptible, whereas PEH-associated strains were more likely to be ciprofloxacin-resistant but TMP-sulfa-susceptible. Local authorities then released guidance to recommend specific population-based therapies for empiric treatment of shigellosis. This strategy of using local epidemiology and resistance data to guide management led to effective empiric therapy for nearly 70% of patients despite a high prevalence of multidrug resistance. Given the risk of continued community transmission of Shigella in MSM with sexually acquired shigellosis, we recommend treatment using a risk and local susceptibility-guided approach.

Notwithstanding the ample literature documenting sexual transmission of Shigella from a clinical and epidemiologic standpoint, knowledge and practice gaps among MSM, healthcare providers, and public health professionals remain. Surveys and focus groups of MSM have revealed that awareness of shigellosis is low, even among individuals with high awareness of other STI (208, 209). While greater knowledge may increase the willingness of MSM to implement certain interventions, like cleaning hands or genitals before sex, recommendations to use barrier protection during rimming, fingering, or fisting may be considered infeasible or outside the norm of some community practices (210). Healthcare providers also lack awareness that Shigella may be sexually transmitted. A U.S. survey conducted in 2020 found that only 18% of healthcare providers identified sexual transmission as a route of Shigella acquisition, even though MSM were identified as an at-risk group by 35% of respondents (211). Several practice gaps relating to the control of sexually transmitted Shigella also exist within the public health field. In 2021, the U.S. CDC began recommending that public health officials collect a sexual history from individuals with shigellosis to improve data quality and aid in outbreak detection (212). Some state health departments were already collecting these data. However, as of 2022, only half of states routinely asked adults with shigellosis about their sexual history, and the data elements collected were variable and often incomplete, including some states only asking men without a travel history or failing to ask about female partners (62). Such gaps in knowledge and practice limit the ability to track and prevent transmission.

e. Entamoeba histolytica

The protozoan genus Entamoeba comprises single-cell cyst-forming parasites or commensals of animals. At least seven species of Entamoeba have been found in the human intestinal tract (213, 214). Only one of these, Entamoeba histolytica, is considered a pathogen. WHO estimates that more than 50 million new E. histolytica infections occur each year, mostly in low or middle-income countries including India, Mexico, and many countries in Africa and Latin America (215, 216). E. histolytica infections are acquired by ingestion of contaminated food or water, or from fecal-oral contact during sexual activity. Gastrointestinal symptoms of amebiasis range from asymptomatic carriage (most commonly) to diarrhea or dysentery. A localized inflammatory colonic mass referred to as an “ameboma” may mimic malignancy (217). Men are more susceptible to amebiasis, possibly due to effects of androgens on innate immunity (218–220).

Intestinal E. histolytica infections typically have a subacute onset over 1-3 weeks. Common symptoms are diarrhea, abdominal pain, and bloody stools; dysentery is characterized by the presence of blood and mucus in stool (221). Extraintestinal infections most notably include amebic liver abscess, with less frequent involvement of lung, heart, and brain. The presentation of amebic liver abscess may be acute or chronic, with right upper quadrant abdominal pain, fever, and other constitutional symptoms. Chronic presentations may be mistaken for malignancy (222).

E. histolytica infections are diagnosed by microscopy, PCR, antigen testing, culture, or serology. PCR or antigen-based assays are recommended over microscopy for the diagnosis of amebiasis in MSM due to higher sensitivity and specificity (223). Microscopy cannot distinguish E. histolytica from E. dispar or E. moshkovskii. However, a Gal/GalNAc lectin specific for E. histolytica can be detected by an antigen test (224). Serologic tests are highly sensitive for the diagnosis of amebic liver abscesses, although they may be negative in the first week of illness (2) or may be positive as a result of prior illness.

Amebiasis is treated with metronidazole or tinidazole. As nitroimidazoles are not reliably cysticidal, paromomycin is administered to kill luminal cysts (225). Treatment of asymptomatic E. histolytica infection in MSM is recommended to prevent the subsequent development of symptomatic disease (226) or transmission to others. Amebic liver abscesses usually respond favorably to medical therapy but may require drainage if very large (>10 cm), unresponsive to medical therapy, or considered to be at imminent risk for rupture, although evidence for the benefits of aspiration is inconclusive (227). Aspiration is also performed to exclude the possibility of a pyogenic liver abscess. Microscopy, PCR or antigen testing may demonstrate the presence of E. histolytica in abscess fluid (228, 229).

The parasitologist Peter Sargeaunt classified strains into “zymodemes” on the basis of multi-enzyme gel electrophoresis and noted that specific zymodemes seemed to be non-pathogenic (230). A landmark study showed that 20% of 225 MSM carried what appeared to be E. histolytica in stool based on parasite morphology, but when isoenzyme electrophoretic analysis was performed on 34 isolates, all were zymodemes I or III, which were considered non-pathogenic (231). Non-pathogenic zymodeme type I was subsequently determined to be a non-pathogenic species called E. dispar (213, 232). The present consensus remains that E. dispar is non-pathogenic, although occasional symptoms and even liver abscess have been associated with E. dispar (233). Additional species E. moshkovskii and E. bangladeshi have also been described (214, 234, 235), but their pathogenicity is not presently known. E. moshkovskii was nearly as abundant as E. dispar in a prospective survey of MSM in Australia (236). Notably, these species are morphologically identical by light microscopy. Although E. histolytica exhibits more efficient phagocytosis of red blood cells, the presence of internalized erythrocytes is not pathognomonic for E. histolytica (237). Some commercial multiplex PCR panels detect E. histolytica, but they may also detect E. dispar if the organism load is sufficient (238). If the distinction between E. histolytica or a non-pathogenic amoeba remains in doubt, an E. histolytica antigen test or colonic biopsy should be obtained.

Sexually transmitted amebiasis was described in MSM as early as 1968, with a cluster of cases in NYC; the index case had traveled from India (2). Some have reported that extraintestinal amebiasis may be more frequent in patients with HIV/AIDS (239). Outbreaks of amebiasis in MSM have been reported from Canada, the U.S. and Spain (4, 57, 240, 241). Oral-anal contact, anilingus and fellatio after anal-genital intercourse are associated with amebiasis in MSM (223, 242–246); direct rectal inoculation during receptive anal sex can also occur.

Early reports of widespread amebiasis in MSM failed to distinguish between E. histolytica and non-pathogenic E. dispar. In fact, when zymodeme analysis or species identification has been performed, non-pathogenic species have been far more predominant (231, 236, 247–250). Nevertheless, invasive amebiasis caused by E. histolytica including colitis and liver abscess has been documented in MSM [reviewed in (223)]; it is unclear whether HIV coinfection is a risk factor for more severe disease or mortality. Although E. dispar is considered to be nonpathogenic, diarrhea, abdominal pain, bloating, and nausea have been reported in some MSM carrying E. dispar in whom no other enteric pathogens were detected (251); some have received antiparasitic treatment, although it is not known whether such treatment was beneficial.

f. Enterobius vermicularis

Enterobius vermicularis (formerly Oxyuris vermicularis) is a parasitic nematode also known as “pinworm” (due to the female worm’s pointed tail) or “threadworm” (which has also been used to refer to S. stercoralis) (252). Humans are the only host of E. vermicularis, which is found worldwide. The global incidence and prevalence of enterobiasis is not well characterized but may vary by age and region, with most recognized infections occurring among children (253, 254). Historically, pinworm has been considered one of the most common helminth infections (252). Infection occurs via ingestion of eggs from other infected individuals, environmental sources, or self-inoculation. Pinworm eggs hatch into larvae in the small bowel and mature into adult worms in the cecum. At night, pregnant female pinworms migrate to the anus, where they deposit eggs in perianal skin folds. Sexual transmission has been reported in case reports dating back to 1972 (255), but it is infrequently identified.

Most cases of E. vermicularis infection are asymptomatic. The most common presentation is pruritis ani, which may be worse at night. This can be accompanied by abdominal pain, anorexia, and sleep disturbances. Pinworm can also rarely present as vulvovaginitis (256, 257), although this is uncommon. E. vermicularis has been associated with appendicitis, but a causal relationship remains uncertain. Additional unusual presentations reported in MSM include eosinophilic ileitis, colitis and buttock folliculitis (258–260). Rarely, sexually transmitted infection can result in acute-onset urethritis following insertive anal sex (261–263). However, the most common presentations are asymptomatic infection or perianal irritation (264). Diagnosis is made by identification of eggs or mature worms on microscopy. Specimens may be collected using a tape test with application of clear tape to the perianal area to collect eggs in the morning, prior to bathing or toileting, or with the use of a commercially available paddle (265). In some instances, worms may be directly visualized. Treatment with albendazole, pyrantel pamoate, or mebendazole given in two doses two weeks apart is generally effective, but reinfection is common. Case reports of sexual transmission of E. vermicularis have implicated asymptomatic partners as sources of infection (264), emphasizing the importance of testing or empirically treating all sexual partners of identified cases.

Sexually transmitted E. vermicularis is often identified along with other STI and STEI. Prior studies have reported HPV, syphilis, Chlamydia, N. gonorrhoeae, Giardia, and E. histolytica in association with pinworm infection in MSM (56, 263, 266–268). However, most cohort studies of individuals attending STI clinics have not identified pinworm. This is likely due to a high prevalence of asymptomatic infections, low sensitivity of stool microscopy for diagnosis, infrequent use of tape or paddle testing for infection, and the lack of a molecular diagnostic test. A cohort study of MSM performed in the late 1970s in Scotland reported that 11 of 118 men (9%) had either pinworm eggs or mature worms identified (267), but more contemporary studies have not been performed.

g. Giardia duodenalis

Giardia duodenalis, previously known as Giardia lamblia or Giardia intestinalis, is a flagellated protozoan organism with two life stages-- a trophozoite present in the intestine of the host and an environmentally stable infectious cyst excreted in feces. Giardiasis, or symptomatic infection by G. duodenalis, is marked by diarrhea, cramping, bloating, flatulence and sulfuric belching. Nausea and vomiting are less common. Traditionally, the diagnosis is made by visualization of the cyst or trophozoite in stool via microscopy. However, this method is labor-intensive and time consuming as it requires examination of three separate stools over multiple days to achieve 90% sensitivity. Modern diagnostics include antigen testing, immunofluorescence, and PCR-based assays. These tests, available since the late 1990s, are highly sensitive and specific and require considerably less time and training to perform.

Giardia was one of the first organisms to be described as sexually transmitted. As early as 1972, a letter in response to an earlier article in JAMA asserted that the presence of multiple intestinal pathogens in a young man might have been the result of sexual transmission rather than ingestion in contaminated food (255). In 1977, a case series of MSM in a single sexual network in Seattle suggested a chain of transmission (269). Three papers referencing a cohort of men participating in the Gay Men’s Health Project in New York City were published between 1977-1979 and calculated a point prevalence of giardiasis ranging from 12-18% (19, 242, 270, 271). Behavioral data from the same cohort revealed that 86% of the men participated in anilingus during the past year (242).

A 1980 comparative study from Toronto used a prospective cohort design to measure relative rates of E. histolytica and Giardia infection among 200 MSM and 100 MSW with or without symptoms of gastroenteritis. The authors found that 13% of MSM had Giardia, in comparison to only 3% of MSW (272). The presence of an intestinal pathogen was only associated with lack of cleansing before anal sex, but not with foreign travel, living in a “gay household,” or having >10 sexual partners in the last 6 months. Additional cross-sectional and point prevalence studies from the early 1980s showed variable rates of Giardia among MSM (139, 243, 244, 267, 273–281). Many found rates around 3%-7%, but a pair of studies from Sweden among MSM attending STI clinics showed rates as high as 9-10% (275, 282).

One of the first studies to evaluate enteric pathogens in the setting of HIV/AIDS was conducted in 1988 and compared MSM with or without symptomatic enteritis in Baltimore (157). Sixty-eight participants with acute diarrhea or proctitis were compared with a subset of 250 asymptomatic participants in a longitudinal study of MSM who did not have HIV/AIDS at enrollment and 77 MSM with AIDS who were recruited as inpatients. Symptomatic men without HIV had Giardia in 13.4% of cases, while all others had similar recovery rates of Giardia ranging from 3-5%. A smaller cohort study from Canada similarly showed little correlation between HIV status and presence of Giardia (283). Perhaps anticipating sexual network analyses, a 1997 study of PWH showed Giardia present in 17.7% of patients using both microscopy and immunofluorescence techniques (27). A high rate of geographic clustering of cases was noted, along with a much higher prevalence among MSM compared to non-MSM (OR 14.2).

Culture-independent diagnostics have likely increased the detection of Giardia in symptomatic patients. A 2020 evaluation of MSM presenting with gastroenteritis in Seattle found 20.4% positivity for Giardia and no impact of HIV status (22). A 2019 study of MSM with gastrointestinal symptoms presenting to an STI clinic in Spain found that 11% were positive for Giardia using antigen detection assays (158). On the other hand, a study of asymptomatic MSM in Melbourne found that only 8 of 591 (1.4%) evaluated by multiplex PCR were positive for Giardia (107), similar to a study of asymptomatic MSM in Taiwan, among whom 8 of 375 (2.1%) tested positive by multiplex PCR (77). Randomized clinical trials have concluded that tinidazole is the most effective treatment for giardiasis, with higher rates of parasitological cure compared to metronidazole or albendazole (284). Unfortunately, refractory giardiasis appears to be increasingly common, particularly in patients who are severely immunocompromised or have traveled to Asia (285) and may require treatment with alternative agents, such as the combination of tinidazole with albendazole or mebendazole, or quinacrine (286).

h. Strongyloides stercoralis

Strongyloides stercoralis is a parasitic nematode endemic to tropical and subtropical regions with a rising presence in temperate regions (287). Humans and dogs serve as definitive hosts as part of a complex life cycle with both parasitic and free-living stages. People acquire infection through penetration of the skin by filariform larvae, although strongyloidiasis can also be transmitted from person-to-person. Strongyloidiasis is considered a neglected tropical disease with an estimated 600 million people globally who are chronically infected (288), but the true burden is uncertain because of limited surveillance and underdiagnosis (289). Light microscopy of stool or duodenal contents is commonly used but may miss chronic infections with low or intermittent larval shedding. ELISA assays targeting S. stercoralis antigens and PCR assays have also been developed (290).

Acute infection with S. stercoralis is characterized by a rash at the site of penetration, which may be followed 1-4 weeks later by abdominal pain, diarrhea, and pulmonary symptoms (289, 291). Larvae migrate to the lungs via veins and lymphatics. In the lungs, they enter the bronchial system and eventually ascend to the pharynx, from which they are swallowed. In the small intestine, larvae mature and produce eggs that release rhabditiform larvae, which are shed in stool to enter a free-living state or can directly reinfect the host, leading to chronic infection. Chronic infection may have subtle symptoms, such as nonspecific abdominal pain and diarrhea, or can be asymptomatic in up to half of cases (288, 292). Immunocompromised individuals are at risk of fatal hyperinfection or disseminated disease with abdominal pain, pneumonitis, sepsis, and meningitis. Therapeutic options for the treatment of acute or chronic disease consists of ivermectin or albendazole.

Data regarding sexual transmission of S. stercoralis are limited but suggest it can occur. The earliest reported cases include a study of STEI from 1981 in New York City, which found S. stercoralis in both MSM and MSW attending an STD clinic (243). One case was identified out of 64 MSW tested and two cases among 51 MSM. No cases were identified in the 48 MSMW included. Significant risk factors for helminth infection were MSM status and oral-anal contact. International travel was associated with a lower risk of helminth detection. Similarly, a case report from 1983 reported S. stercoralis in an MSM with no travel history who had sexual contact with an asymptomatically infected man with recent travel to Colombia and prior residence in Brazil (293). A 1984 report of MSM in London identified one S. stercoralis infection out of 363 gay men with gastroenteritis and no cases among asymptomatic controls (294). A recent systematic review that included some of these as well as additional case reports identified risk factors for sexually-transmitted S. stercoralis including oral-anal contact, STI, HIV, other protozoal infections, travel to an S. stercoralis endemic country, having multiple sexual partners from endemic areas, and engaging in chemsex (295, 296).

i. Cryptosporidium spp.

Cryptosporidium is a genus of parasitic protozoans that infects humans and a wide range of other vertebrates. Globally, Cryptosporidium spp. account for 10.2% of all global deaths from diarrheal disease, with many occurring in children under 5 years of age or with severe immunosuppression (297). The first two case reports of human cryptosporidiosis were published in 1976; one was a 3-year-old child and the other an immunosuppressed adult (298, 299). Cryptosporidium hominis and C. parvum cause most human cryptosporidiosis (300). Infection is acquired through ingestion of oocysts, which replicate in the apical portion of the intestinal epithelium and are intermittently shed in stool. Oocysts are stable in soil and aquatic environments and relatively resistant to chlorination (301). Transmission occurs via contaminated food or water, including swimming pools, as well as direct person-to-person spread or contact with infected animals. Diagnosis can be made by light microscopy of stool samples, electron microscopy of small bowel biopsy specimens or most commonly, by molecular methods. Cryptosporidium is included in some commercially available multiplex PCR panels.

Cryptosporidiosis in humans was rarely identified in the pre-AIDS era, and there is overlap between the first recognized cases of cryptosporidiosis and potentially undiagnosed HIV. A case series from 1981 summarized six published cases of Cryptosporidium-induced diarrhea, of which one from 1980 was a gay man with recent travel to Mexico who also had disseminated CMV, although a formal immunodeficiency diagnosis was not made (302). A report of 21 men with AIDS and cryptosporidiosis was published by the CDC in 1982 (303). Of the 21, twenty were MSM. As well-documented in early case reports and case-series, cryptosporidiosis in immunocompromised adults causes life-threatening chronic watery diarrhea. Cryptosporidium spp. can also infect immunocompetent adults, in whom it causes self-limited diarrheal illness lasting from a few days to weeks and resolving without treatment (304, 305). Other symptoms can include fever, fatigue, abdominal pain, and proctitis (108). Some infections may be asymptomatic. Treatment with nitazoxanide is FDA-approved (306). However, nitazoxanide is less effective in malnourished children or people with advanced HIV and was not found to be more effective than placebo in a meta-analysis of studies in PWH (307). This underscores the need for novel antiparasitic agents (308) and the importance of global access to ART.

The environmental stability of Cryptosporidium spp. makes it difficult to definitively link cases to sexual transmission as opposed to indirect transmission through contaminated surfaces or water. However, starting with early case reports of cryptosporidiosis in MSM, there have been documented instances of men without travel or animal exposure who became infected, and clustering among MSM with or without AIDS, which has inferred sexual transmission (309, 310). A rigorous analysis of cryptosporidiosis in MSM published in 1986 reported an 8% prevalence in symptomatic MSM compared to 2% in the general population (p=0.017) (140). A 1994 study of people with AIDS in Los Angeles found cryptosporidiosis to be significantly more common among people for whom sexual contact was their primary risk factor for HIV acquisition (30). Serology-based studies also support sexual transmission of Cryptosporidium, with recent sexual activity, anal sex, and greater numbers of sexual partners all significantly associated with prior exposure to the parasite (25).

Inclusion in multiplex PCR panels has increased the diagnosis of cryptosporidiosis in the last decade (311). Nevertheless, it remains strongly associated with advanced HIV, and prevalence inversely correlates with CD4 count (22, 312). A recent systematic review of cryptosporidiosis risk factors among MSM found that in addition to HIV positivity, younger age, African-American race, and prior infection with Treponema pallidum or E. histolytica were demographic factors associated with infection (313). Behavioral risk factors included having ≥10 sexual partners in the prior 2 years, visiting sex-on-premises spas or saunas, and recreational drug use. Younger age may be related to risk due to either behavioral factors or a lack of protective antibodies (314). Racial disparities in infection rates are likely related to social determinants of health.

j. Cystoisospora belli

Cystisospora belli (formerly Isospora belli) is a single-celled intracellular coccidian parasite that infects the epithelium of the small intestine (315). This protozoan is named for its bell-shaped oocysts, though this appearance is inconsistent (316). C. belli is transmitted through ingestion of oocysts, which are shed intermittently in feces (317) and can persist in water or food. Humans are the primary host of C. belli, although recent studies of other mammals have identified novel genotypes in other species, suggesting potential zoonotic transmission (318). C. belli is traditionally diagnosed by microscopy, with visualization of oocysts in stool or sporozoites seen on intestinal biopsy or in duodenal aspirate. More recently, PCR-based diagnostics have been developed to detect C. belli (319), and it is included as a target on some multiplex panels (320).

C. belli causes persistent diarrhea and is most frequently identified in people in tropical or subtropical regions or with severe immunosuppression. Most cases are mild and can involve large-volume diarrhea, nausea, anorexia, headache, fever and abdominal pain. Some infections may be asymptomatic. Cases assumed to be acquired via sexual contact have the same presentation as those acquired through other means. In PWH with a CD4+ cell count <200 cells/μl, organ transplant recipients, and people receiving chemotherapy or with other forms of immunosuppression, C. belli can cause severe acute or chronic diarrhea, biliary infections, and disseminated disease. The incubation period for C. belli in immunocompetent hosts is 7-10 days (315, 317). Oocysts are subsequently excreted for 2-5 weeks following the onset of symptoms. Symptoms and shedding resolve spontaneously, although treatment may shorten the duration of illness. TMP-sulfa is the recommended treatment, but immunosuppressed individuals may experience relapse and require repeated courses (321).

Cases of C. belli infection have been reported among MSM without a travel history or other risk factors for acquisition, suggesting that it is sexually transmitted (322). The earliest English language report of C. belli among MSM was a case series from New York describing three men diagnosed between 1980 and 1981 (323). One patient had traveled to Mexico and Guatemala, and all had multiple sexual contacts. It is unclear whether the men were HIV-positive at the time of infection, but one went on to be diagnosed with AIDS in 1983, and the other two were lost to follow-up. A similar case series from Los Angeles identified three MSM with C. belli enteritis and no travel history (324). Based on T cell subset ratios, all three likely were immunocompromised, presumably from HIV. In both case series, coinfection with other protozoa was seen in some individuals (323, 324). These studies mark the beginning of a rise in symptomatic cases of C. belli in the U.S. since the early 1980s, which was driven by the enlarged pool of susceptible immunocompromised adults with HIV. Subsequent studies of MSM in which C. belli has been identified are limited, suggesting varying prevalence of C. belli driven by regional networks and the degree of HIV-related immunosuppression in the population or differences in pathogen detection. A 1986 study of MSM in Chicago, Illinois with diarrhea and unknown HIV status found that 2 out of 372 samples were positive for C. belli (0.7%), making it one of the least commonly detected parasites in their study (325). Similar results were reported in 1988 in Seattle, where C. belli was not detected in asymptomatic MSM with or without AIDS and found in only 1 of 45 (2.2%) of MSM with AIDS and diarrhea (157). In contrast, a study of MSM with HIV conducted in the mid-1990s in Los Angeles reported 10% positivity for C. belli (326), although this study is exceptional, and more recent studies of MSM in Seattle and New York City using PCR for testing did not identify any cases (22, 100).

k. Microsporidia

Microsporidia are a diverse and ubiquitous group of spore-forming parasitic organisms that evolved from a common ancestor with fungi (327–331). Microsporidia are of clinical importance as opportunistic intracellular pathogens in immunocompromised hosts (332). The most frequently identified microsporidian pathogen infecting humans is Enterocytozoon bieneusi (333). Like many microsporidia, E. bieneusi spores are environmentally stable and lead to infection when ingested; spores inject sporoplasm into host cells via polar tubules, followed by proliferation in the cytosol and the formation of new spores that are released through cell lysis to be shed in stool. Detection can be made by light microscopy of stained fecal samples, electron microscopy of intestinal biopsy specimens, or PCR (334). Serologic testing can assess prior exposure (335).

Other microsporidia known to infect humans and reported in association with HIV include Anncaliia (syn. Brachiola) vesicularum, Encephalitozoon cuniculi, Encephalitozoon hellem, Encephalitozoon intestinalis, Pleistophora ronneafiei, Trachipleistophora anthropopthera, Trachipleistophora hominis, Tubulinosema acridophagus, and Vittaforma corneae (331). Although person-to-person transmission has not been investigated for most microsporidial species, some are transmitted primarily as zoonoses, by arthropod bites, or through exposure to contaminated water or food. Microsporidiosis is typically a gastrointestinal infection causing either acute or chronic diarrhea (21), but some microsporidia may cause disseminated disease or infect extraintestinal tissues including the cornea, paranasal sinuses, respiratory tract, brain, liver, biliary tract and skeletal muscle (336). Some microsporidia may respond to albendazole or fumagillin treatment, although reversal of the underlying immunosuppressive condition should be attempted when possible (337).

Microsporidiosis has been associated with sexual transmission since its initial descriptions in MSM with symptoms of AIDS, and the association between symptomatic microsporidiosis and HIV-related immunosuppression is strong. However, the link to sexual transmission has not been made as conclusively as for other sexually transmitted enteric pathogens. A serologic study of 30 MSM with symptoms of AIDS in Sweden published in 1984 found that 33% had antibodies to E. cuniculi (335), compared with 12% serologic positivity among 115 Swedish adults from the general population with recent travel to tropical regions (338). Other studies from the pre-antiretroviral therapy (ART) era also identified high rates of prior or current microsporidial infection in MSM. A 1990 study using electron microscopy to evaluate intestinal biopsies found that 30% of MSM with AIDS and chronic diarrhea in whom prior microbiological testing was negative were infected with E. bieneusi (339). Prevalence of microsporidiosis among people with HIV is strongly and inversely associated with CD4 count (340, 341), and the introduction of ART has been associated with near eradication of cases among people with HIV in high income settings (342). Some studies have inferred that microsporidia may be sexually transmitted based on higher prevalence in MSM (23, 31, 343). However, a weakness of some of these studies is that the relationship between sexual activity and microsporidiosis may be confounded by CD4 count and environmental exposure. The data most strongly suggesting possible sexual transmission come from a case-control study of people with a CD4 count ≤200 cells/mm³ and E. bieneusi or E. intestinalis and unmatched controls with a CD4 count ≤200 cells/mm³, which found that MSM status was associated with microsporidiosis after controlling for age, area of residence, CD4 count ≤100/mm³, and swimming in a pool in the prior 12 months, with an adjusted odds ratio of 7.6 and 95% confidence interval of 1.0 – 59.4 (23). While there may be sexual transmission between immunosuppressed individuals, the ubiquity of microsporidia in aquatic environments makes this only one of many possible routes of infection.

l. Hepatitis A

Hepatitis A virus (HAV) is a highly infectious non-enveloped RNA virus that infects the liver. Globally, HAV is a common cause of acute liver injury, with 159 million cases estimated in 2019 (344). Humans are the only host for HAV, which is transmitted via a fecal-oral route either through direct person-to-person contact or from contaminated water or food. Sexual transmission of HAV has long been recognized (4, 345–347), facilitated by relatively long periods of shedding before and after clinical illness. In immunocompetent adults, viral shedding in stool precedes clinical symptoms by 1-2 weeks and continues for 3 months or more after clinical recovery (348, 349). Immunocompromised individuals may shed virus for longer periods (350).

Clinically, HAV infection has identical presentations in both sexually transmitted and non-sexually-transmitted cases. Common symptoms include fatigue, dark urine, jaundice, anorexia, nausea, and abdominal pain (351). Approximately half of adults with HAV infection report diarrhea. Treatment consists of supportive care. Sexually transmitted cases in MSM are associated with hospitalization rates of 40-50% (351, 352). Death is rare, with no deaths reported in most sexually-transmitted outbreaks, and liver failure reported in less than 1% of cases (353). Laboratory studies are typically notable for elevated aspartate aminotransferase and alanine aminotransferase levels, often exceeding 1,000 U/L. The presentation is indistinguishable from other causes of acute viral hepatitis. Diagnosis is primarily made by serology, with anti-HAV IgM used to identify acute infection (354). Because the presentation of acute hepatitis from HAV is similar to other forms of acute viral hepatitis, additional testing for HBV and HCV is typically obtained to rule out coinfection. HAV can be detected using molecular methods, and occasionally PCR is used to detect viremia or the presence of HAV in stool. Acute HAV infection is a CDC reportable condition.

Vaccination against HAV is highly effective, with nearly 100% seroconversion rates in immunocompetent adults. Although vaccination is still effective for adults with HIV or other forms of immunosuppression, response rates vary (355–357). HAV vaccination has been recommended for MSM in the U.S. since 1995 (358). The WHO also recommends vaccination of MSM as one of several specific groups at elevated risk for infection (344). Despite WHO’s guidance, there is variability in recommendations for HAV vaccination of MSM across Europe, and the costs of vaccination vary (352, 359). Fewer than 50% of European MSM report ever having been vaccinated against HAV, and knowledge regarding viral hepatitis and awareness of vaccines is highly variable (359, 360). In England, Australia, and the U.S., low rates of vaccine coverage among MSM have been similarly reported (361–363). However, low coverage is not necessarily indicative of widespread vaccine refusal. Studies show that MSM seeking PrEP have high HAV vaccine uptake when it is included in clinical care (364), as do MSM receiving routine primary care (365). A cost-effectiveness analysis of English data found that proactive vaccination of MSM is cost saving relative to reactive vaccination during an outbreak or no vaccination, although data from other countries have shown that reactive vaccination campaigns can still reduce the number of total cases and shorten outbreaks (366, 367). For unvaccinated individuals exposed to HAV, postexposure vaccination is recommended as soon as possible within two weeks (368). For people who have chronic liver disease or are immunocompromised and unvaccinated, administration of HAV immunoglobulin is recommended along with HAV vaccination within two weeks of exposure. Epidemiologic modeling indicates that vaccination rates of >70% among MSM are necessary to prevent sustained transmission (369).

In MSM, HAV outbreaks have been infrequent but sometimes extensive. From June 2016 through May 2017, a multi-strain outbreak involving 1,400 confirmed cases, some of which were linked to the EuroPride festival (370), occurred across 17 European countries, of which 84% of cases were MSM and 92% were unvaccinated (352). Overall, 45% of cases required hospitalization. Three HAV strains were identified, with evidence that they were genetically related to the initial strain detected, indicating repeated transmission and viral evolution over the year-long outbreak period. The outbreak continued until 2019, and outbreak strains were also detected in many other countries, including Canada, Israel, the U.S., Chile, Japan, and Taiwan (370–375). Closely related HAV genotype 1A strains were subsequently implicated in outbreaks in Brazil and Croatia among MSM (376, 377), implying local spread after initial introduction from the European outbreak. Recently, one of the outbreak strains has been found to be circulating in MSM in Portugal (378). It is uncertain whether this represents continued low-level transmission or reintroduction.

The high infectious potential of HAV allows outbreaks to be perpetuated through multiple modes of transmission. HAV outbreaks among MSM may initially arise from an environmental source but then spread further through sexual networks. One example of this phenomenon was seen in a 2022 outbreak of HAV that predominantly spread among MSM but was linked by sequencing data to an initial source of contaminated frozen berries (379). Overall, 171 individuals were affected, of whom 82% were male, some with concurrently diagnosed bacterial STI. HAV outbreaks spreading from person-to-person in the MSM community can also cross over into non-MSM groups. The 2016-2017 EuroPride-related HAV strains also infected women and adult men without reported MSM behavior, presumably through person-to-person or environmental contamination, such as in bathrooms or other environments (380). In the Netherlands, where population-level HAV immunity was low among adults, 30% of outbreak-related cases were in non-MSM (380).

2. Enteric pathogens with uncertain sexual transmission

a. Clostridioides difficile

Clostridioides (formerly Clostridium) difficile is a spore-forming toxin-producing anaerobic Gram-positive bacterium that causes diarrhea in the setting of gut microbiome disruption. Hospital-acquired C. difficile infections (CDI) are more common among older, antibiotic-exposed, and immunocompromised individuals. Individuals harboring C. difficile may exhibit a spectrum of clinical presentations ranging from asymptomatic colonization to diarrhea or pseudomembranous colitis. In contrast to the aforementioned pathogens, MSM populations typically have lower rates of CDI than general populations undergoing testing for gastroenteritis.

Studies in the pre-molecular testing era reported low rates of C. difficile as a cause of symptomatic diarrhea or proctitis in MSM. A small 1986 study of MSM in Seattle with fewer than four weeks of GI or anorectal symptoms reported only a single C. difficile-positive case (4%) and one C. difficile-positive control (17%, or 1 of 6 controls tested) (381). Subsequent studies have generally reported similar rates of CDI among symptomatic individuals and lower rates among asymptomatic controls. A 1988 study of MSM from the Baltimore area found low rates of CDI among MSM. Of individuals who had diarrhea or proctitis, 5% (6 of 110) tested positive for C. difficile using selective culture and cytotoxicity testing (157). All asymptomatic men tested negative for C. difficile (n=271), including those with advanced HIV. A similar National Institutes of Health study from the late 1980s involving 30 MSM with advanced HIV (20 with diarrhea and 10 without) did not detect any C. difficile cases by cytotoxicity assay (382). No C. difficile infections were detected in a 1999 Swiss study of PWH (both men and women) with acute diarrhea, and a low prevalence of C. difficile (2.7-3.7%) was found in PWH with chronic diarrhea (21).

In the molecular testing era, C. difficile rates among symptomatic MSM have remained low. A 2019 study of MSM in Seattle reported that 6% of individuals were positive by multiplex PCR assay for C. difficile, substantially lower than the 12.4% positivity rate identified in persons tested from the general population (22, 53). Similarly, a study of MSM and non-MSM adults with HIV in New York City who underwent GI PCR testing found that rates of C. difficile detection were low (1-2%) in both MSM and non-MSM, with no significant difference between groups (100). A rectal swab-based study in Australia detected C. difficile in only 1 of 499 MSM with proctitis and no cases among asymptomatic MSM (n=506) (108). The consistently lower rate of CDI in MSM across studies may be attributable in part to differences in key sociodemographic factors between MSM being tested versus other adults who are often older, more likely to be female, and reporting recent international travel.

Low rates of C. difficile among MSM have remained stable over time, even as molecular testing has improved detection of C. difficile (383) and community-acquired cases have become more common (384). This is surprising, as sexually active MSM are considered a highly antibiotic-exposed population, yet their risk of C. difficile acquisition does not appear to be elevated compared to the general population. The unexpectedly low rates of CDI may be due to the types of antibiotics to which MSM are exposed, enhanced resilience of their gut microbiome, or other unidentified host immune factors. STI treatment typically involves low intensity regimens (i.e. narrower spectrum and/or shorter duration), and low antibiotic treatment intensity is associated with a lower risk of CDI (385). Some studies have also reported that MSM have richer and more diverse gut microbiomes than non-MSM (386, 387), which may confer resistance against C. difficile. Specific differences in the gut microbiome observed in MSM include enrichment of Prevotellaceae, including the Segatella copri complex (388). However, the potential relevance of these differences to host susceptibility to enteric pathogens remains to be established. Lastly, MSM included in studies of STEI tend to be younger (<50 years old) and have fewer comorbidities, which may reduce their risk of CDI, although patients with community-acquired CDI also tend to be younger with fewer co-morbidities than those with hospital-acquired CDI (389, 390).

a. Yersinia spp.

Yersinia enterocolitica and Yersinia pseudotuberculosis are enteric Gram-negative bacteria that cause abdominal pain, diarrhea, vomiting, fever, mesenteric adenitis, and inflammation of the terminal ileum. Yersiniosis is often implicated in sporadic outbreaks linked to contaminated food. The lack of evidence linking Yersinia to sexual transmission may be due to a variety of factors, including acquisition primarily as a zoonosis with negligible person-to-person spread, low prevalence in high-resource settings, and difficulty detecting Yersinia spp. prior to the advent of multiplex PCR testing.

Pigs are the primary reservoir for Yersinia, although the bacteria are also found in other domesticated and wild animals. Handling or consuming undercooked pork or raw dairy products are common exposure routes. Humans can shed Yersinia spp. for months following infection, but person-to-person transmission is rare (391). Overall, the rate of yersiniosis in the U.S. is declining (392). Outbreaks among MSM have not been described in the literature, and sporadic cases have not been linked to sexual networks. Studies have infrequently detected Yersinia among MSM with diarrhea (22). Among asymptomatic MSM, the reported prevalence varies: 1.7% of rectal swabs from Australian men were positive for Yersinia spp. (107), but a similar study in New York did not detect any cases (105). In the general population, rates of yersiniosis rates do not differ significantly by sex (393).

Molecular panels have significantly increased the detection rate of Yersinia spp. in the United States because traditional culture and isolation require special handling to maximize recovery. These techniques include cold enrichment, alkali treatment, or plating on CIN (cefsulodin-irgasan-novobiocin) agar (394). Molecular tests for Yersinia target Yersinia enterocolitica but do not distinguish between pathogenic and nonpathogenic biovar 1A strains. Fluroquinolones are the preferred treatment (395).

b. Intestinal spirochetosis

Intestinal spirochetosis is an unusual condition distinct from syphilitic proctitis, which is characterized by the overgrowth of anaerobic spirochetes (Brachyspira pilosicoli, B. aalborgi) along the luminal surface of the colonic mucosa (396, 397). This condition is most frequently encountered in MSM, in particular PWH, leading to the suggestion that it may be sexually transmitted (398, 399), but sexual transmission has not yet been directly demonstrated. A recent study analyzed 165 patients with human intestinal spirochetosis in Spain; 87% were MSM, and 81% were symptomatic (most with diarrhea). Symptomatic improvement was noted following treatment with metronidazole or doxycycline (400). Symptoms of intestinal spirochetosis may include diarrhea, abdominal pain, constitutional symptoms, flatulence, and proctitis (rectal pain, bleeding, discharge, tenesmus). Diagnosis is made by biopsy and based on morphologic criteria; the presence of spirochetal organisms appears as a “pseudo-brush border” that can be visualized by Warthin-Starry staining (401).

c. Enteric viral pathogens

Sexual transmission of enteric viruses such as norovirus, rotavirus, astrovirus, adenovirus, and sapovirus is conceivable and perhaps likely but has not been well established. Many studies of gastrointestinal infections have not assessed the presence of viruses; when viruses have been detected, their association with sexual activity in comparison to other enteric pathogens has not been evaluated.

Early studies from the late 1980s were inconsistent regarding the prevalence of viral enteric pathogens in MSM. An Australian study from 1988 using culture, ELISA, and microscopy-based identification methods reported viral pathogens other than cytomegalovirus in 54% of stool samples of hospitalized MSM with HIV and diarrhea (402). Most of the viruses identified were rotaviruses and adenoviruses. In the control group of asymptomatic MSM with or without HIV who presented for STI testing, enteric viruses were detected in approximately 15% of cases, with no difference in prevalence by HIV status. There may have been regional or seasonal variation in viral pathogens, as a study conducted by the U.S. National Institutes of Health in MSM with HIV and diarrhea failed to detect rotavirus infections among the 20 adults evaluated (382). Neither study included non-MSM control groups to ascertain the community prevalence of viral enteric pathogens.

More recent use of multiplex PCR assays has allowed the greater recognition of viral pathogens in MSM with gastroenteritis. Two U.S. studies suggest that norovirus, rotavirus, and sapovirus may be present at higher rates among MSM with gastroenteritis in comparison to the general population. In the late 2010s, norovirus was more commonly detected among MSM than non-MSM with gastroenteritis in Seattle (22); compared with 3.3% of all tests overall detecting norovirus and 1.7% detecting sapovirus, 15.2% of MSM tested positive for norovirus and 5.3% for sapovirus. At around the same time, a study of PWH with gastroenteritis in New York City found that MSM were more likely to have a GI pathogen detected by PCR compared with non-MSM (100). Although the differences failed to achieve statistical significance, 20.6% of MSM had a viral pathogen detected compared with 12.9% of non-MSM. Most of the viruses detected were noroviruses, with small percentages of enteric adenovirus, astrovirus, rotavirus, or sapovirus.

Among asymptomatic MSM and MSM presenting with isolated proctitis, detection of enteric viruses may be less common. An anorectal swab-based study of Australian MSM with or without proctitis found low rates of norovirus (0.6% in proctitis cases and 0.2% in controls without proctitis), rotavirus (0 cases among proctitis cases and 0.6% among controls), and sapovirus (0.4% among proctitis cases and 0 among controls) (108). A similar study, also using anorectal swabs, was conducted in New York City among MSM receiving HIV PrEP (105). In that cohort, 1% of tests were positive for norovirus and 0.5% positive for sapovirus. Individuals reported GI symptoms in all cases in which viruses were detected. A third study from Melbourne identified viral pathogens in 2.9% of rectal swabs from asymptomatic MSM attending an STI clinic from 2018-2019 (107). Most of the viruses detected were astrovirus (1.9% of patients tested), with lower rates of norovirus (0.4%) and rotavirus (0.6%). In that study, detection of any enteric pathogen was more common among men reporting group sex or rimming, but the authors did not analyze whether these factors were associated with specific pathogen types.

Overall, enteric viruses are detected at low rates from rectal swabs of asymptomatic MSM. Whether these rates are low because asymptomatic individuals are unlikely to be infected or because rectal swabs are an insensitive test sample is uncertain. It is also unclear whether detection of enteric viruses in MSM is related to sexual transmission or other routes of exposure. Asymptomatic norovirus infection prevalence in the general North American adult population outside of an outbreak is unknown. Norovirus prevalence among North American infants and children is approximately 4% (403), but adult prevalence is likely to be lower, as a 2004 study of Australian adults detected no norovirus in asymptomatic adults (404), and a 2016 cross-sectional study of 4,536 adults in Japan detected norovirus in 2.5% (405). Nevertheless, asymptomatic viral shedding may occur for weeks after norovirus infection (406).

d. Hepatitis E

Hepatitis E virus (HEV) is a non-enveloped single-stranded RNA virus. Of the different genotypes of HEV, HEV-1 and HEV-2 only infect humans and are typically found in low-resource settings, where fecal-oral transmission occurs from person-to-person contact or via contaminated water or food. HEV-3 and HEV-4 are more common in high-resource settings and infect a range of mammalian hosts, including game animals. Infection with these genotypes is usually acquired through consumption of raw or undercooked meat or offal. HEV is shed in the stool. Diagnosis is typically made by a serum HEV IgM assay or HEV RNA detection by PCR. Epidemiologically, HEV in both low- and high-resource settings has a 2 to >3:1 male: female ratio for infections in the general population (407). The reasons for this skew are unclear, but the age distribution differs from what is typically seen in STEI, with HEV generally infecting older adults (aged >60).

Because of the similar transmission pathway to HAV, there has been speculation that HEV may be sexually transmitted. Serologic studies have provided mixed results but suggest that sexual transmission is uncommon. One study suggesting the possibility of sexual transmission analyzed 977 serum samples from MSM and 141 heterosexual male controls from the UK collected between 2006 and 2008 and found MSM significantly more likely to have positive HEV IgG than controls from the same clinics (odds ratio 3.1, p = 0.025). Prevalence in MSM increased over the study period from 2.1% in 2006 to 9.3% in 2008 (408). HIV status was not associated with HEV seropositivity. An Italian study conducted in 2015-2017 found that HEV IgG positivity was more common among MSM seen in an STI clinic compared to male blood donor controls, with 10.2% of MSM having positive serology versus 5.2% of controls (p<0.05) (409). MSM were also more likely to be both HAV and HEV IgG-positive (14%) compared to controls (1%). However, studies suggesting that HEV is not sexually transmitted are more numerous. A French serosurvey of people with HIV conducted in 1995 among 100 consecutive clinic patients found no cases of HEV seropositivity among people with HIV acquired through sexual contact (410). A Taiwanese study of 3,293 people with HIV (87% MSM) from 2015-2017 found that HEV prevalence was 3.7% and incidence was 4.35 per 1,000 person-years of follow-up (PYFU), which were significantly lower than the prevalence and incidence of HAV infection during the same period (31.1% and 12.61 per 1,000 PYFU, respectively) (411). Sexual orientation and sexually transmitted infections were not associated with HEV incidence or prevalence, suggesting that it was not sexually transmitted. Similarly, data from a study in the Netherlands found that HEV seroprevalence in MSM was the same as in blood donors (24%) and not associated with higher numbers of sexual partners in the prior six months or with HIV status (412). A study of adults in Spain with acute HAV between 2017 and 2018 measured HEV IgG seroprevalence (413), finding that MSM were more likely to be admitted with acute HAV than non-MSM, but both groups had the same seroprevalence of HEV IgG (7%). No incident HEV, based on serum HEV RNA PCR, was observed in either group. Lastly, data from a substudy within the ANRS IPERGAY trial in France found that HEV IgG seroprevalence was 14.9% and not associated with number of sexual partners, bacterial STI, recreational drug use, or HAV status (414).

5. Emergent Themes in Sexually Transmitted Enteric Infections

Although STEI have been recognized for more than half a century, several emergent themes are noteworthy. First, the development of molecular culture-independent testing methods has improved the ability to diagnose these infections and impacted empiric management of STEI. Second, the increasing incidence of MDR Gram-negative bacteria is creating therapeutic challenges and warrants improved surveillance across various geographical and socioeconomic settings. Third, although appreciation of the role of sexual transmission of enteric pathogens in MSM has improved over time, gaps in our understanding of the epidemiology of these infections remain for many organisms. Changing transmission dynamics over time have now been well documented for Shigella, by nature of its infectiousness and tendency to cause symptoms, and such studies point the way for the study of other STEI. The importance of the gut microbiome in host susceptibility to STEI and the potential impact of doxycycline used in doxy-PEP for bacterial STEI prevention remain to be defined. Lastly, improvements in public health approaches to prevent these infections are sorely needed.

a. Changes in Laboratory Diagnosis

Several studies have examined the impact of multiplex molecular diagnostics in MSM presenting with acute gastroenteritis. Some enteric infections, such as cryptosporidiosis, have increased in incidence following the introduction of multiplex panels (311). Multiplex panels have likely contributed to this increase, both by increasing analytical sensitivity and by detecting infections that were not clinically suspected. It has also become clear that the spectrum of enteric pathogens in MSM is broader than originally appreciated.

Sobczyk et al. compared multiplex PCR panels with conventional stool testing in 1,941 specimens obtained between 2013 and 2017 in persons with HIV (415). One-hundred twenty-four samples were positive by multiplex PCR, compared with only 45 by conventional testing. Co-infections were more commonly detected by PCR than by conventional workup (48.4% vs 13.3%), and the more rapid turnaround time (mean 23.4 vs 71.4 hours) allowed the avoidance of unnecessary antibacterial therapy in 9 of 11 individuals who had only viral pathogens detected. Conventional ova and parasites examination detected some nonpathogenic parasites.

Cybulski et al. performed a retrospective cohort study of 235 MSM who underwent multiplex PCR testing for acute gastroenteritis at one of two Seattle medical centers from 2017-2018 (53). Fifty-eight percent of the patients had a pathogen detected by a large multiplex panel. Forty-three percent of the positive samples had more than one pathogen detected, which was rarely encountered prior to the introduction of multiplex diagnostic tests. The organisms most frequently detected were diarrheagenic E. coli (predominantly EAEC and EPEC), Shigella spp., Campylobacter spp., Giardia and norovirus. The increased detection of EAEC, EPEC and norovirus is unsurprising, as these pathogens are not found by traditional diagnostic methods. The increased detection of Shigella may in part reflect suboptimal survival of Shigella in Cary-Blair holding medium, as well as the presence of enteroinvasive E. coli, which carries the Shigella virulence plasmid and is detected by molecular tests for Shigella but not recovered by selective media for Shigella. Reflexive cultures for Shigella and Campylobacter spp. revealed a high prevalence of multidrug resistance, including to beta-lactams, macrolides, fluoroquinolones, and sulfonamides. At the same institutions, Miller et al. subsequently performed a retrospective analysis of fecal multiplex PCR results obtained from 361 MSM in 2019 (61). Sixty-two percent of the samples were positive for a pathogen, with no significant difference observed between individuals with or without HIV. Persons with diarrhea who have poorly controlled HIV and negative multiplex panel results are more likely to require hospitalization (100).

More recent studies have examined the prevalence of enteric pathogens in asymptomatic MSM. Overall prevalence in asymptomatic MSM is estimated to be 1.1% for Shigella, 1.9% for Campylobacter, 3.8% for diarrheagenic E. coli, and 0.3% for Salmonella (148). Enteric pathogens are more likely to be detected in individuals using PrEP, living with HIV, having >5 new sexual partners in the past 3 months, or participating in insertive oral-anal sex or group sex (106). A convenience sample of rectal swab specimens from patients with rectal Chlamydia infections in the UK found that 8.6% were positive for an enteric bacterial pathogen (104). Forty-seven percent of the patients carrying an enteric pathogen lacked gastrointestinal symptoms. Using a 15-target multiplex panel, investigators in Melbourne documented asymptomatic carriage of pathogenic enteric organisms in 11% of asymptomatic MSM (107). These results are comparable to a cross-sectional study of rectal swabs from 2,116 MSM attending a London sexual health clinic, in which a multiplex PCR assay detected bacterial pathogens in 9.8% of participants (106). In addition, 41.3% of individuals with bacterial pathogens detected were positive for the mphA gene, a marker of macrolide resistance. Only 1.7% of the overall cohort reported gastrointestinal symptoms. Collectively, these data demonstrate that asymptomatic carriage of enteric pathogens is not uncommon in MSM and helps to account for the failure of symptom-based diagnosis to prevent ongoing transmission.

At a NYC sexual health clinic, Purpura et al. performed a longitudinal cohort study of 110 MSM who were taking HIV PrEP (pre-exposure prophylaxis) (105). Forty-five percent of the participants had at least 1 pathogen detected by multiplex PCR, and some tested positive for the same pathogen over a period of months. Although gastrointestinal symptoms were common among the participants, symptoms were not significantly associated with the detection of a pathogen by PCR. A survey of anal samples from 212 asymptomatic MSM attending a sexual health clinic in Amsterdam found Shigella in 2.8%, with higher rates in MSM who had recent or active gastrointestinal symptoms (147)

In a study of 375 fecal samples from asymptomatic MSM in Taiwan, recent use of proton pump inhibitors (PPIs) was significantly more frequent in those containing at least one enteric pathogen (22.2% vs 2.0%, P < 0.001) (77). This identifies a potentially modifiable risk factor for STEI. The use of PPIs has previously been shown to increase the risk of community-acquired enteric infections (416), most likely because of the microbicidal activity of gastric acid and pH-dependent reactive nitrogen species (417)

In summary, multiplex PCR panels have had a major impact on the diagnosis of gastroenteritis in MSM. The breadth of targets has allowed the identification of unsuspected pathogens and revealed the increased incidence of mixed infections. Detection of Salmonella, Shigella, Campylobacter or STEC should be followed by reflexive culture to recover isolates for susceptibility testing and public health tracking (418). The spectrum of enteric pathogens in MSM is now established to include diarrheagenic E. coli and norovirus, which are not detected by conventional stool workups. Improved analytical sensitivity in comparison to traditional methods has improved diagnostic accuracy. Moreover, rapid diagnostic turnaround time and the detection of common viral pathogens has facilitated more targeted antibiotic use (53, 178). The specificity of multiplex panels for most pathogens is high, although cross-reactivity between E. histolytica and E. dispar remains an issue and may require resolution by the more specific E. histolytica antigen test. Importantly, multiplex panels are cost-effective, with the modestly higher reagent costs offset by improvements in lab workflow and other resource utilization (54).

b. Rising Incidence of AMR

In 2019, the CDC stated that the world had entered a “post-antibiotic era” characterized by widespread resistance, illness and death related to multi-drug-resistant organisms (MDRO) (40). This was not an entirely unexpected announcement given that multiple enteric pathogens have developed antimicrobial resistance at increasing rates over the past 30 years. Reasons for this global trend have been well summarized elsewhere (419–423). Factors particularly relevant to understanding patterns of MDR and XDR STEI among MSM include a rise in international travel, changes in sexual behavior, and the widespread usage of antimicrobials [Figure 1].

Antimicrobial resistance of sexually transmitted enteric pathogens has been noted since the 1970s (167), when tetracycline-resistant Shigella transmission was found in MSM in Seattle. Resistance to ampicillin and clindamycin was detected in Campylobacter fetus strains isolated from MSM in the 1980s (424). In 1989, a Campylobacter cinaedi (now Helicobacter cinaedi) isolate from an HIV+ MSM in Australia was reported to have resistance to erythromycin and norfloxacin (425), and two additional isolates had norfloxacin resistance. In a S. sonnei outbreak in San Francisco in 2000, predominantly involving MSM, 19 of 20 randomly selected isolates were resistant to TMP-sulfa, tetracycline, ampicillin, and streptomycin, although they were susceptible to ciprofloxacin, nalidixic acid, and 3^rd generation cephalosporins (33). The rise of resistant organisms has continued to accelerate with increased global travel facilitating the widespread dissemination of resistant clones. Importantly, the treatment options for XDR enteric pathogens are extremely limited (205, 426).

Among enteric pathogens with established sexual transmission, Campylobacter, Salmonella, and Shigella are all currently identified as serious drug-resistance threats. From the 1990s to 2017, the percentage of Campylobacter isolates with decreased susceptibility to ciprofloxacin nearly doubled, rising from approximately 15% to nearly 30% of isolates tested, with 2% also harboring resistance to azithromycin (40). Nontyphoidal Salmonella has exhibited increasing resistance to ceftriaxone, ciprofloxacin, and azithromycin, with 16% of isolates in 2017 harboring resistance to at least one antibiotic (40) Drug-resistant Shigella infections underwent exponential growth, with a 5- to 10-fold increase in the percentage of cases with resistance to ciprofloxacin or azithromycin between 2009 and 2017; in 2017, 3% of cases were estimated to be resistant to both ciprofloxacin and azithromycin (40). An analysis of Shigella flexneri serotype 2a isolates from 2016-2018 in Ontario found that approximately half were azithromycin- or ciprofloxacin-resistant, and 35% of the azithromycin-resistant isolates were also ciprofloxacin-resistant; clusters predominantly involved men and were caused by strains carrying an IncFII plasmid that has been associated with global outbreaks among MSM (427).

In 2022, a pan-resistant Campylobacter infection was detected in an MSM coinfected with Mpox (44). The Campylobacter infection spontaneously cleared, but nevertheless heightened concern over increasing antimicrobial resistance in Campylobacter spp., for which treatment options are already limited. A 2022 retrospective study of MSM with HIV in Italy reported that 75% of Campylobacter isolates were resistant to fluoroquinolones, 50% were resistant to tetracyclines, and 17% were resistant to macrolides (45).

XDR Shigella strains, defined as resistant to azithromycin, ciprofloxacin, ceftriaxone, TMP-sulfa, and ampicillin, have been reported in the US and Europe since 2016 (46, 205, 428) and more recently in Australia (198). As of 2022, XDR strains were increasingly common and represented 5% of all isolates tested in the U.S. XDR infections are more likely to be seen among MSM, PEH, international travelers, immunocompromised people, and people with HIV. XDR Shigella sonnei resistant to ampicillin, TMP-sulfa, ciprofloxacin, azithromycin and ceftriaxone was reported in eight MSM following a LGBTQ+ festival in Chicago (429); the strains were clonally related and carried bla_CTX-M-27 along with 10 other resistance genes. XDR Shigella sonnei resistant to ampicillin, TMP-sulfa, ciprofloxacin, azithromycin and ceftriaxone was recently reported in three MSM in Los Angeles, California (430); sequence analysis demonstrated the presence of bla_CTX-M-15.

Mechanisms driving the emergence and persistence of MDR and XDR STEI among MSM include extended duration of infection or frequent reinfection, global travel to endemic regions, carriage of antimicrobial resistance genes on low fitness-cost plasmids, antimicrobial use for treatment of other STI, and MDR pathogen transmission among international sexual networks. Long-term carriage or frequent reinfection of MSM with circulating STEI creates an opportunity for the development of MDR organisms. A study of an outbreak in the UK found identical isolates from the same patient taken over 4 months apart (102), and individuals may remain symptomatic with the same STEC serotype for over a year (431). Similarly, adults infected with Shigella can become long-term carriers following infection, with viable bacteria in the gut for up to 17 months (188). Sexual networks of MSM have also changed. MSM are now reporting more sexual partners than in the past (432) and less consistent condom use than during the 1990s and 2000s (433, 434). There are multiple contributors to these changes, including the uptake of HIV PrEP and use of geolocating mobile phone apps for meeting partners (435, 436). The result is expanded opportunities for STEI transmission and larger networks for spread.

Travel from regions with high rates of resistance is an important contributor to MDRO spread. As many as 81% of Salmonella isolates from China are resistant to at least two classes of antimicrobials, compared with just 2.2% of US isolates, a pattern likely driven by local behavioral and ecological factors (437). One example demonstrating how international travel can introduce new resistant strains is the spread of MDR S. flexneri among MSM and PEH in Seattle from 2018-2022. Genomic analysis revealed that a strain with ampicillin and ciprofloxacin resistance was acquired via an unknown route during travel to Southeast Asia then circulated in the US among MSM, likely through sexual transmission, before spilling over into PEH, among whom limited access to sanitation may have been a primary driver of ongoing transmission (178). International travel can also link sexual networks, allowing greater spread of MDR organisms. A retrospective study from 2020 identified nearly identical strains of MDR Campylobacter coli in MSM from Seattle and Montreal between 2015 and 2018, suggestive of an individual or individuals with sexual contacts in both cities (75).

AMR genes conferring resistance with low fitness costs have been identified in multiple pathogen types. Horizontal gene flow is likely to exist between Shigella and closely-related pathogenic E. coli strains, which can co-circulate among MSM communities (102), with closely-related plasmids carrying shared AMR genes detected in outbreaks of STEC and Shigella involving MSM in the UK. MDR Shigella resistomes are predominantly associated with conjugative plasmids that collectively carry nearly all potential antibiotic resistance genes required for resistance to first-line drugs for shigellosis (195). Shigella sublineages can share AMR determinants, and an outbreak in the UK demonstrated shared resistance genes carried on the plasmid pKSR100 in co-circulating strains of S. sonnei, S. flexneri 2a, and S. flexneri 2b (438, 439). Since 2020, pKSR100-type plasmids have acquired the extended-spectrum beta lactamase (ESBL) gene bla_CTX-M-27, leading to an increase in XDR ESBL-producing S. sonnei in MSM (48). Similar plasmids have been sporadically identified in S. flexneri isolates from MSM in similar geographic regions, suggesting possible horizontal transfer between co-circulating species. After an initial appearance in MSM in high-income countries, MDR plasmids and genomic islands are now found in Shigella isolates from LMICs as well (440, 441).

Although the absolute rate of antimicrobial use among MSM with STEI is not known, prior studies have found that MSM with STEI often have recent or concurrent bacterial STI (442, 443), which may be treated with antimicrobials such as azithromycin or ceftriaxone (42). This creates community pressure for antibiotic resistance, especially in STEI with long duration of carriage or frequent reinfection. MSM seeking STI testing also have high rates of inappropriate antibiotic use. A retrospective study in Sydney found that MSM unnecessarily received ceftriaxone in 50% of visits for anorectal symptoms (444). Low-specificity presumptive treatment algorithms for MSM at risk for STI also increase unnecessary antibiotic use. A study of the WHO presumptive treatment algorithm for Neisseria gonorrhoeae and Chlamydia trachomatis in asymptomatic MSM in Kenya found that only 25% of men eligible for treatment with cefixime were positive for either pathogen by nucleic acid amplification testing (445).

The gut microbiome is a reservoir for antibiotic resistance genes, with many enteric pathogens readily able to share genetic information (446). A study of PWH in Spain reported that although MSM do not have a greater abundance of antimicrobial resistance genes in their gut microbiome, they carry a more diverse array of resistance genes that are distinct from those found in non-MSM (447).

c. Potential Impact of Doxy-PEP

Doxycycline post-exposure prophylaxis (doxy-PEP) has been found effective in multiple clinical trials for prevention of bacterial sexually transmitted infections (STI), including Chlamydia, syphilis, and to a lesser extent, gonorrhea, in MSM and transgender women (448). Real-world doxy-PEP implementation in Northern California has shown decreased rates of syphilis and Chlamydia infections, with mixed results for gonorrhea (449, 450). Meanwhile, a rising prevalence of tetM-carrying Neisseria gonorrhoeae has been observed in the U.S. as doxycycline is increasingly used for the treatment and prevention of STI (451). Although the U.S. CDC endorses doxy-PEP for these populations in individuals with a history of bacterial STI in the past year, the guidelines also acknowledge that doxy-PEP use carries the potential risk for development of AMR and alteration of the gut microbiome (452). This warning is particularly salient given that MSM experience a disproportionate risk for AMR based on increased antimicrobial use and MDRO exposure within sexual networks (453).

Exposure to doxycycline, a broad-spectrum antimicrobial in the tetracycline (TCN) class, is known to induce the expression of tetracycline resistance genes (TRG) in vitro as well as in clinical settings (454, 455). TCN exposure is associated with decreased numbers of health promoting commensal Enterobacteriaceae, Enterococcus spp., E. coli, Fusobacterium spp., and Streptococcus spp. (456, 457). While these risks have not yet been well characterized in the setting of doxycycline use as STI prophylaxis, concerns have been sufficient to lead some international organizations to advocate for either restricted use or against doxy-PEP altogether pending further data on its long-term impacts (458).

Outside of the STI field, chronic doxycycline use is not uncommon and has been employed for decades in various settings across the world for malaria prophylaxis, treatment or prevention of dermatologic conditions like acne or rosacea, and treatment of periodontal disease. Many studies have sought to evaluate the impact of chronic doxycycline use for non-STI prophylaxis on the GI microbiome, but drawing unified conclusions from these studies has proven challenging due to variability in the populations being evaluated, the type of tetracycline and dose used, and differences in analytic methods (bacterial culture vs 16S rRNA gene sequencing, etc.). While data on AMR in the setting of doxycycline for STI prophylaxis, including doxy-PEP and doxycycline pre-exposure prophylaxis (doxy-PrEP), are emerging, primary outcomes of interest for real-world implementation are similar to those in prior studies of doxycycline use for non-STI prophylaxis. These include the timing and extent of tetracycline resistance (a surrogate for doxycycline susceptibility), changes to bacterial diversity in the gut or other anatomic sites, and emergence of resistance in off-target bacteria.

A recent systematic review examined the impact of oral tetracyclines on AMR from isolates collected at various anatomic sites and concluded that use led to an increase in tetracycline resistance in the GI tract but not resistance to other antibiotic classes. The heterogeneity of these studies did not allow performance of a meta-analysis (459). One in vitro study exposed fecal suspensions collected from 3 healthy volunteers to various quantities of tetracycline and evaluated changes in the GI microbiome at 24 hours and 40 days. Tetracycline-treated samples had either negligible or minor changes in the total number of 16S rRNA gene copies, slightly increased abundance of Bacteroides and Clostridium family XI, and a variable but increased abundance of TRGs (460). Another small study in which 2 volunteers took 200 mg of doxycycline per day for 7 days found increased relative DNA abundance and higher expression of TRGs compared to controls (461). Interestingly, post-doxycycline treatment samples also had higher numbers of enteric species with bacteriophages containing tet family-specific elements with the potential for horizontal gene transfer (461). Whether by conjugation or transduction, horizontal gene transfer is a well-established mechanism of expanding tetracycline resistance in Bacterioides spp. and other Gram-negative bacteria (462).

In one randomized double-blind placebo trial, daily doxycycline use vs. placebo for 3 weeks was associated with a higher proportion of resistant commensal E. coli isolates (100% vs 53% resistance to tetracycline) in stool, which returned to baseline levels after 2 weeks off doxycycline (463). Older studies also described a return to pre-antibiotic levels one month after doxycycline exposure (456, 457). Another RCT that evaluated E. coli resistance patterns from stool isolates of acne patients taking tetracycline or erythromycin and their family members found tetracyclines to be associated with significantly higher levels of tetracycline resistance as well resistance to agents in other classes such as sulfonamides and ampicillin compared to the erythromycin-exposed cohort (464).

Data on the occurrence of AMR collected from non-STI doxycycline prophylaxis users is otherwise largely observational. One study comparing doxycycline to mefloquine prophylaxis did not find either agent to be significantly associated with increased risk for bacterial enteropathogens such as Campylobacter or more resistant non-ETEC compared to incidence in local Thai soldiers not using malaria prophylaxis (465). Another study found that healthy U.S. Army soldiers deployed to Afghanistan who had limited exposure to the environment outside their military base and were using daily doxycycline for malaria prophylaxis had a 5.5-fold higher rate of community-onset MDR E. coli colonization compared to healthy recruits living stateside; 90% of isolates had tet(A) or tet(B) genes detected, consistent with doxycycline resistance (466).

To date there is scant evidence to support a causal relationship between intermittent doxycycline exposure for STI PEP and deleterious alterations in the GI microbiota, although this has been hypothesized (467, 468). Recent observational data from two large efficacy trials provide some insight into possible impacts on the GI microbiome specific to doxy-PEP use. Investigators in the French ANRS 174 DOXYVAC study found no difference in the prevalence of ESBL E. coli from anorectal swabs of participants in either control or doxy-PEP groups (range: 31.9-38.7%) over the course of 18 months of follow-up (469). The US DoxyPEP trial used DNA-seq and RNA-seq to evaluate self-collected rectal swabs from participants receiving doxy-PEP or control subjects over time. TRGs comprised nearly 50% of the resistome by mass at enrollment, although <5% were expressed resistance genes. There were no significant differences between arms in the resistome mass, alpha diversity, or beta diversity by DNAseq from baseline to 6 months, but the number of expressed TRGs as measured by RNAseq increased from 4% to 15% in the doxy-PEP group by 6 months (470). The proportion and richness of expressed TRGs was highest in participants who took more than 25 doses of doxycycline over the study period, suggesting a significant dose-response relationship. Notably, this study did not find that the proportion of resistance genes from other antibiotic classes increased in the doxy-PEP group over time. Consistent with a prior systematic review and meta-analysis (471), doxycycline use was not associated with an increased risk of C. difficile infection. Finally, an open-label, randomized pilot trial conducted in Vancouver assigned MSM and transgender women taking HIV PrEP to immediate vs. deferred use (beginning at 24 weeks of follow-up) of doxy-PrEP (taken as 100mg of doxycycline daily) and evaluated the presence and changes in microbial DNA from rectal swabs at baseline, 24 weeks and 48 weeks using 16s rRNA sequencing (472). Of 41 participants with samples analyzed at all time points, doxy-PrEP had a minimal impact on the rectal microbiome over the 48-week study period. Beta diversity did not differ within or between study arms over time; alpha diversity was similar within participants in the two arms at the genus and family level. Abundance of Fusobacterium decreased in both arms by 1.7% at 48 weeks, but this estimate did not pass multiple hypothesis testing (473) and thus remains of unknown clinical significance.

Overall, evidence is mixed but confirms that intermittent doxycycline exposure for any purpose can increase the abundance of transcriptionally active TRGs, which may in turn result in clinical tetracycline resistance and reduce the numbers of some prevalent and important commensal bacteria in the gut microbiome. When resistance emerges, it seems to be specific to the tet family and spares other important classes like beta-lactams (459), although some studies have found an effect on other antibiotic classes as well (464). While limited by a relatively short follow-up period, evidence from recent doxy-PEP trials and one doxy-PrEP trial suggest that the intervention does not lead to multi-class AMR nor cause significant changes in the diversity or composition of the microbiome or increase the prevalence of MDR E. coli (469, 470, 473). Nevertheless, doxycycline use is associated with increased expression of TRGs in the short term, potentially due to a combined effect of selective pressure, suppression of susceptible bacteria, and the expansion of TRGs already present in non-susceptible bacteria (470). The clinical significance of these findings and potential implications for clinical practice and public health remain unknown at this time. As doxycycline for STI prophylaxis is implemented in the United States and beyond, comprehensive evaluations of AMR will be critical and should be prioritized in diverse settings and populations to gain a fuller understanding of its impact on the gut microbiome.

d. Importance of Sexual Practices

Many risk factors for enteric infections in MSM have been identified [Table 3, Figure 2]. These include HIV (including a low CD4 count and the absence of antiretroviral therapy), numbers of sex partners, residence in urban MSM districts, PrEP use, AMR strains, concomitant STI or protozoal infections, current or prior enteric infections, travel to endemic areas, use of dating apps, and behavioral factors including condomless receptive anal sex, oral-anal sex (anilingus or rimming), oral-genital sex, penile-anal sex, sex toys, fisting, scat play, multiple or new sexual partners, group sex, casual sex partners, sex-on-premises venues, meeting sex partners via the internet, and chemsex (61, 76, 98, 295, 313). The many behavioral correlates of infection suggest that preventive interventions may be possible.

Table 3.

Sexual Practices Associated with Transmission of Enteric Infections in MSM.

Practice or Risk Factor	Possible Pathogens	References
Oral-anal contact or sex (anilingus, “rimming” or fellatio after genital-anal sex)	Campylobacter	(76)
	E. histolytica	(223, 242–246)
	Shigella	(189)
	Giardia	(242)
	Strongyloides	(243) (295, 296)
	Diarrheagenic E. coli	(1, 98)
Penile-oral or penile-anal contact or sex	E. histolytica
	Campylobacter	(76)
	Diarrheagenic E. coli	(1, 98)
	Cryptosporidium	(25)
	E. vermicularis
Use of sex toys	Diarrheagenic E. coli	(1, 98)
Fecal-oral contact, including digital-anal contact or use of ungloved fingers during sex	E. histolytica	(223, 242–246)
	Shigella	(155)
	Cystoisospora	(317)
	Norovirus
Fisting and/or “scat play”	Shigella	(210)
	Diarrheagenic E. coli	(98)

Figure 2:

Risk factors for sexually transmitted enteric infections among MSM

While the full spectrum of sexual practices has been known for millennia, changes over time may have affected transmission dynamics. It is clear from sexual behavioral data collected over the last 50 years that oral-anal contact is strongly associated with sexual transmission of enteric infections (4, 107), as is greater numbers of sexual partners. These studies found that >70% of MSM presenting with sexually transmitted enteric infections had participated in rimming. Although the precise incidence of STEI in MSM is not known, it is likely that “safe sex” practices aimed at reducing HIV transmission (474) also impacted the epidemiology of STEI. Accordingly, more recent outbreaks of enteric infections in MSM, including multidrug-resistant strains of Shigella and Campylobacter, have occurred as condomless sexual practices were increasingly observed among MSM (475), along with an increase in other STI (476).

As discussed, the impact of pre- and post-exposure prophylactic interventions targeting HIV, gonorrhea and Chlamydia infections (477, 478) on enteric infections is unknown. Nevertheless, there is concern that interventions to reduce the risk of specific STI might lead to more sexual risk-taking, in what has been referred to as “risk compensation” (479, 480). This could in turn facilitate the transmission of enteric pathogens, which are not targeted by the PrEP and Doxy-PEP interventions. However, data from the DoxyPEP trial showed that following a few months of increased numbers of sexual partners, sexual behaviors among participants returned to baseline (481).

In a study from 2015, Gilbart and colleagues conducted semi-structured interviews with 42 MSM diagnosed with Shigella flexneri 3a (482). They found that these men had dense sexual networks with high rates of attendance at sex parties. Chemsex was reported by 50% of participants, and 43% of those felt that the drugs had negatively impacted their decision making. Chemsex is not novel, but data on rates of chemsex and its impact on transmission dynamics were not well understood until the past few decades. The use of social networking and geolocating mobile phone apps to find sex partners is new over the last few decades. In the Gilbart study, these were common modes of finding sex parties and individual sexual partners. Other investigators have found increased number of sexual partners, chemsex, and condomless anal intercourse among those using apps as compared to those not using apps (483). The correlation between number of sex partners, chemsex, and sex party attendance and acquisition of sexually transmitted enteric pathogens suggests that rates of enteric pathogens in MSM may track with rates of other STI. Indeed, British data confirm that from 2010-2014, the rate of STI among MSM rose sharply (484), and circulating S. flexneri 3a rose exponentially during that time (485).

National surveillance data from the U.S. demonstrate that increases in STI among MSM are followed by increases in STI among MSW and cis-women. This has most notably been observed in recent syphilis trends (486). However, because culture-independent testing has had a dramatic impact on population-wide rates of diagnosed Shigella and other reportable STEI, and sexual behavior data are inconsistently collected in enteric pathogen surveillance, it is difficult to ascertain reliable trends over time (62, 487).

Outbreaks of sexually transmitted enteric pathogens have not been reported among heterosexual sexual networks. This may be the result of different rates of insertive oral-anal contact between heterosexual men and women. One study in Australia among people attending a sexual health clinic found that 25.5% of men had performed rimming, in comparison to only 9.3% of women, whereas women were more likely than men to receive rimming (488). This asymmetry in sexual practices may limit the potential for sequential transmission of enteric pathogens. It is also possible that outbreaks of STEI are underrecognized in heterosexual networks because sexual histories are not routinely obtained when patients present with a chief complaint of diarrhea (61). Moreover, sexual history is not reported on most case forms used in the U.S. and United Kingdom for enteric infections that are reported to public health authorities (62).

6. Future Directions

As the epidemic of enteric infections in MSM continues unabated, with recurring outbreaks, increasing levels of AMR, and spill over into other populations, it is increasingly clear that current public health approaches have been inadequate. Specific surveillance of STEI and MDR enteric pathogens at a global level must be improved, which will require better coordination between agencies accustomed to focusing on conventional STI or foodborne pathogens. The British Association of Sexual Health and HIV (BASHH) has recently published guidelines for the treatment of STEI (489). More aggressive use of antimicrobial therapy may reduce the likelihood of ongoing transmission but might also drive the emergence of resistant strains. Surveys indicate that knowledge regarding STEI among MSM is low (209), identifying an opportunity for greater education to reduce risky sexual practices, improve partner notification, and enhance care-seeking for GI symptoms. Interviews with MSM have indicated receptivity to some interventions to prevent shigellosis (e.g., washing hands, genitals and anus prior to sex) but more resistance to others that are perceived to be outside social norms (e.g., use of dental dams for oral-anal contact, latex gloves for fingering or fisting) (210). With rapid and accurate diagnostic tools now available, novel approaches for the control of sexually transmitted enteric infections in MSM should be considered, such as the detection and treatment of asymptomatically colonized individuals (204). It is time to develop a more effective paradigm for STEI control and prevention.

Biographies

Kira L. Newman, MD, PhD is a Clinical Assistant Professor at the University of Michigan. She completed medical training and a PhD in epidemiology at Emory University, internal medicine residency at the University of Washington, and a fellowship in gastroenterology and hepatology at the University of Michigan. She is part of the University of Michigan’s Inflammatory Bowel Disease (IBD) Program, providing care for people with Crohn’s disease, ulcerative colitis, and other forms of IBD. She has research expertise in enteric pathogens, the role of the microbiome in IBD, and digestive health for sexual and gender minority people.

Gretchen Snoeyenbos Newman, MD is an Assistant Professor at Wayne State University (WSU). She is the Director of the WSU Adult HIV Program, which includes Tolan Park Clinic, Detroit’s safety net clinic for people with HIV (PWH) and the Michigan AIDS Education and Training Center. She completed medical school and residency at Emory University and Infectious Disease Fellowship at University of Washington. Her research stems from her clinical work and focuses on implementation of existing and novel models of care for PWH and developing the clinician workforce to address the syndemics of HIV, STIs, and viral hepatitis.

Chase A. Cannon, MD, MPH is an Assistant Professor at the University of Washington (UW) and Medical Director of the Seattle-King County Sexual Health Clinic. He completed medical training at the University of Texas Southwestern, internal medicine residency at Tulane University, and an infectious diseases fellowship with MPH in Epidemiology at the University of Washington. He currently provides primary care for people with or at risk for HIV and sexually transmitted infections (STI) at the UW, including in low barrier care clinics. His area of expertise is sexual health, which broadly includes HIV/STI care and prevention and a special focus on the sexual health needs of men who have sex with men. His research is informed by active clinical practice and focuses on implementation of existing and novel HIV/STI prevention strategies, such as doxycycline postexposure prophylaxis (doxy-PEP), for LGBTQ+ and other socially marginalized populations.

Ferric C. Fang, MD is a Professor of Laboratory Medicine, Pathology and Microbiology at the University of Washington. He is an infectious diseases specialist and molecular microbiologist with over forty years of experience as a clinician, educator and researcher. He obtained his medical training at Harvard Medical School and UC San Diego and previously held a faculty appointment at the University of Colorado. He is presently the Director of Clinical Microbiology at the Harborview Medical Center and has authored more than 300 publications, many focusing on the pathogenesis, epidemiology, diagnosis and management of enteric infections.