Prevalence of Gastrointestinal Pathogens Detected by Multiplex Polymerase Chain Reaction in a Prospective Cohort of Men Who Have Sex With Men Taking Human Immunodeficiency Virus Preexposure Prophylaxis-New York City, 2019-2020.

Multiplex polymerase chain reaction testing for gastrointestinal pathogens was performed on a longitudinal cohort of 110 men who have sex with men taking human immunodeficiency virus preexposure prophylaxis. At least 1 pathogen was detected among 50 (45%) participants, with some participants testing positive for the same pathogen on multiple consecutive visits over a period of months.

Sexual transmission of gastrointestinal (GI) pathogens is both a historical and emerging phenomenon among men who have sex with men (MSM), with outbreaks of Shigella and Campylobacter reported with increasing frequency [1, 2] worldwide. In addition to gastroenteritis, these pathogens can also cause proctitis or proctocolitis, yet current management of rectal symptoms for MSM is empiric treatment for gonorrhea and chlamydia [3] only, and testing for GI pathogens is currently only performed in the setting of diarrhea. In the context of these practices, the burden of antibiotic resistance is increasing in Shigella [4-6] and Campylobacter [2].

Outside of outbreak investigations, the prevalence of enteric pathogens in MSM is not well characterized. Among symptomatic MSM with gastroenteritis in the US state of Washington, Escherichia coli, Shigella, Campylobacter, Norovirus, and Giardia were most frequently identified [7]. Only 1 study, in Melbourne, Australia, has described asymptomatic carriage of pathogenic enteric organisms in MSM, with a cross-sectional prevalence of 11% [8]. To date, prospective data regarding prevalence and symptomatology are lacking. Here we present the prevalence of GI pathogens and correlation with clinical factors in a prospective cohort of MSM receiving preexposure prophylaxis (PrEP) for HIV in New York City.

METHODS

Patient Population

This study was performed at a sexual health clinic within a large urban academic medical center in northern Manhattan. Adult HIV-negative MSM were recruited from a prospective PrEP adherence study (Stick2PrEP) from 7 February 2019 until 12 March 2020, and the enrolled population reflects the overall clinic population [9].

Study Design

At regular 3-month clinic visits for PrEP follow-up, regardless of symptoms, an additional flocculated rectal swab placed in Carey-Blair transport media (FecalSwab, Copan Diagnostics, Murietta, California) was collected at the same time as a clinical rectal swab for gonorrhea and chlamydia testing; the additional rectal swabs were tested for 22 organisms by multiplex polymerase chain reaction (PCR) off-label with the FilmArray GI Panel (BioFire Diagnostics, Salt Lake City, Utah) [10]. GI Pathogen PCR results were used for research purposes only and not made available to patients or their providers. Participants completed an enrollment survey describing demographics and sexual risk factors, and at each visit a survey on GI symptoms and potential exposures. Study follow-up was 12 months but ended early for the majority due to the coronavirus disease 2019 (COVID-19) pandemic.

Data Analysis

Demographics, clinical variables, and pathogen species were assessed using descriptive statistics. Associations with a positive swab, clinical outcomes, and sexual health behavior were assessed using χ 2 for categorical and Student t test for continuous variables. A P value <.05 was considered significant. Statistical analyses were performed using SAS version 9.4 software (SAS Institute, Cary, North Carolina).

Ethics Statement

This study was approved by the Columbia University Irving Medical Center Institutional Review Board.

RESULTS

From 7 February 2019 to 12 March 2020, 110 of 148 (74%) participants in Stick2PrEP study enrolled in the GI pathogen substudy. Overall, 98% identified as male and 45% as Hispanic. Thirty-five percent reported income <$20 000 per year, 45% did not have a college degree or higher, and 45% reported not having health insurance (Supplementary Table 1).

A total of 194 rectal swabs were collected from 110 participants. More than 1 study visit was completed by 50 (45%) participants, with 22 (20%) completing 3 visits, 11 (10%) completing 4 visits, and 1 (1%) completing 5 visits. Median number of days between study visits was 98 (range, 12–315). At least 1 organism was detected in 62 (32%) rectal swabs from 50 (45%) participants; 1 swab was invalid and excluded from the analysis. Enteropathogenic E coli (EPEC) was detected in 20 (10%) rectal swabs, Giardia lamblia in 19 (10%), enteroaggregative E coli (EAEC) in 16 (8%), Campylobacter in 14 (7%), Shigella/enteroinvasive E coli in 4 (2%), enterotoxigenic E coli in 1 (<1%), Vibrio in 2 (1%), Cryptosporidium in 2 (1%), Norovirus in 2 (1%), and Sapovirus in 1 (<1%) (Figure 1, Supplementary Table 2). Giardia was detected in 3 consecutive swabs in 2 participants and 2 consecutive swabs in 1 participant. Campylobacter was detected on 3 consecutive swabs in 1 participant and 2 consecutive swabs in 1 additional participant. EPEC was detected on 3 consecutive swabs in 1 participant and EAEC on 2 consecutive swabs in 1 participant. The median number of days between visits was 91 (range, 28–119) for 2 consecutive positive swabs and 91 (range, 63–119) for a positive swab followed by a negative swab. More than 1 organism was detected in 16 rectal swabs, with 4 organisms detected in 1 swab, 3 in 2 swabs, and 2 in 13 swabs.

Figure 1.

Longitudinal multiplex polymerase chain reaction results of rectal swabs from men who have sex with men receiving human immunodeficiency virus preexposure prophylaxis. Abbreviations: EAEC, enteroaggregative Escherichia coli; EIEC, enteroinvasive Escherichia coli; EPEC, enteropathogenic Escherichia coli; ETEC, enterotoxigenic Escherichia coli; MSM, men who have sex with men; PCR, polymerase chain reaction; PrEP, preexposure prophylaxis; STEC, Shiga toxin–producing Escherichia coli.

Symptoms reported within 30 days of rectal swab collection are reported in Supplementary Table 3. Overall, frequency of symptoms was high irrespective of swab results; 41 of 62 (66%) positive and 81 of 130 (62%) negative rectal swabs were associated with ≥1 symptom. The most commonly reported symptoms were diarrhea (43%), abdominal discomfort (32%), urgency (27%), tenesmus (18%), nausea (18%), and mucus in stool (10%). There was no significant difference in reporting of associated symptoms or self-reported antibiotic use with PCR results. At least 1 symptom was associated with most pathogens; however, only 60% of EPEC, 63% of EAEC, 50% of Shigella, 64% of Campylobacter, and 74% of Giardia positive swabs were associated with at least 1 symptom.

Risk factors associated with acquiring a GI pathogen are displayed in Supplementary Tables 4 and 5. PCR results were not significantly associated with travel, sick contacts, or reported sexually transmitted infection in the preceding 30 days. There was a trend toward less frequent condom use associated with positive swabs compared to negative swabs (4.4 compared to 5.4 condom use out of 10 sex encounters, respectively; P = .06). Although there was no difference in number of sexual partners in the preceding 30 days or perceived sexually transmitted infection risk reported at baseline, a negative swab result was associated with higher, but not statistically significant, number of receptive anal sex encounters in the past 6 months (8.58 vs 5.25; P = .07).

DISCUSSION

To our knowledge, this is the largest longitudinal study to date of enteric pathogen prevalence among MSM attending a sexual health clinic. We observed a surprisingly high detection rate, with 45% of participants having at least 1 positive swab during the study period and 32% of all swabs detecting at least 1 pathogen. Multiple pathogens were detected in 16 of 62 (26%) of swabs. The same pathogen was detected on consecutive swabs from 9 (8%) participants, suggesting long-term carriage. Overall, GI symptoms were common but not significantly associated with PCR results, with 44% of positive swabs not associated with any symptom.

Notably, the distribution of pathogens among MSM in our study differed substantially from the baseline distribution of GI pathogens among all 9402 patients with diarrhea who underwent FilmArray GI Panel testing at our medical center from 2015 to 2017 [11]. Whereas E coli subtypes and viruses accounted for 52% and 39% of all positive tests, respectively, among the general population, they accounted for 60% and 5% of positive tests among MSM in our study. In addition, Giardia was identified much more frequently among MSM (10% vs 1.3%).

While giardiasis has been well-described among HIV-positive MSM [12], its prevalence among HIV-negative MSM is not as well understood. In a prospective study of 253 MSM presenting with acute gastroenteritis in Seattle, Washington, Giardia accounted for 14% of positive stool specimens among HIV-positive MSM and 33% of positive stool specimens among HIV-negative MSM [7]. Our study lends further support to the high prevalence of Giardia in the HIV-negative MSM population, and furthermore, provides evidence of long-term carriage. While EPEC, EAEC, and Campylobacter were frequently detected in both studies, Shigella and viral etiologies (particularly Norovirus), were seen much more frequently among the symptomatic Seattle cohort, suggesting asymptomatic carriage of these pathogens is less common.

The prospective nature of our study allowed us to study long-term carriage. In total, 9 participants had a positive swab at >1 consecutive visit. Giardia or Campylobacter were identified at up to 3 consecutive visits in 5 participants and these swabs were not always associated with gastrointestinal symptoms. These findings have important implications regarding asymptomatic carriage and potential transmission risk to sexual partners; however, further research is needed to definitively evaluate true carriage vs reinfection.

Our study is subject to several limitations. First, participants were recruited from a single academic center and the study was restricted to MSM who were enrolled in a cell phone–based PrEP adherence study, so our results may not be generalizable to the general MSM population. Second, due to the COVID-19 pandemic, our overall sample size was limited and only 46% of participants completed >1 study visit, limiting our ability to assess incidence of pathogen detection and longitudinal carriage. Third, the FilmArray GI Panel is not US Food and Drug Administration approved for use with rectal swabs, although the performance of flocculated rectal swabs has been previously shown not to differ from stool [13]. Fourth, sexual health questions were asked at enrollment and may not represent behavior at the time of swab collection. Last, antimicrobial resistance data were not obtained but would be important to assess from a public health perspective, particularly for Shigella and Campylobacter.

CONCLUSIONS

Overall, we found a high rate of enteric pathogen detection among HIV-negative MSM taking PrEP, with 45% of participants having at least 1 positive rectal swab during the study period, most commonly with EPEC, EAEC, Campylobacter, and Giardia. GI symptoms were common among all participants, but not significantly associated with positive PCR results and may not be useful as screening stratification. While the clinical relevance of EPEC and EAEC is less clear, Campylobacter and Giardia are both associated with sexual transmission among MSM, and the high rate of detection we observed may have important public health implications. Testing of partners within a sexual network would be needed, however, to assess true transmission risk and community spread. Importantly, Giardia is not covered by current empiric treatment for proctitis or lymphogranuloma venereum. Furthermore, emerging antimicrobial resistance of Shigella and Campylobacter is a significant concern, and empiric treatment may further select resistant isolates. Given these concerns, sexual health testing strategies should consider the use of multiplex PCR screening for GI pathogens in MSM with GI symptoms, and asymptomatic screening may be beneficial in high-prevalence or outbreak settings.

Supplementary Data

Supplementary materials are available at Open Forum Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.

Click here for additional data file.