Article Text
PDF
Abstract
Objectives Chlamydia trachomatis (CT) and Neisseria gonorrhoeae (GC) are the most commonly reported diseases in the USA, with increasing concern regarding cephalosporin-resistant GC strains and growing discussion of routine male screening and analysing extragenital sites. Hospital emergency departments (EDs) may be appropriate screening venues, and our objective was to identify the extent of genital and oropharyngeal CT/GC infection in ED patients.
Methods From June 2012 to March 2013, ED patients aged 15–34 were asked to provide a urine sample and oropharyngeal swab for CT/GC analysis and complete a sexual history survey.
Results The 301 female and 192 male participants had a mean age of 25.2 years (SD 4.9); were 65.5% white patients and 33.5% black patients and >85% reported sexual activity in the past year. Tested patient prevalence was 7.7% (any infection/any site; n=38) with no gender differences. Regarding oral infections (n=10), those so infected were more likely to report a friend with a sexually transmitted disease (OR=4.25; CI 1.12 to 16.20), anonymous sex in the past year (OR=5.77; CI 1.58 to 21.15) and belief of some chance of oral infection (OR=5.29; CI 1.31 to 21.28) than those not so infected. Furthermore, four had no corresponding genital infection, and 66.7% (CI 29.1% to 100%) of the oral GC infections were missing concordant genital infection.
Conclusions We find that male and female ED patients have similar likelihood of infection, that 26.3% (CI 12.4% to 40.2%) of those infected have an oral infection and that the majority of oral GC infections would not be identified or treated with urine-based screening. EDs may be important venues to identify those orally infected and provide male screening.
- Chlamydia Trachomatis
- Epidemiology (General)
- Neisseria Gonorrhoea
- Screening
Statistics from Altmetric.com
Background
Despite years of national intervention efforts, Chlamydia trachomatis (CT) and Neisseria gonorrhoeae (GC) remain the two most reported infections in the USA.1 CT infection extent and associated morbidity costs have resulted in the recommendation that sexually active women ≤25 years old be annually screened, but feasibility, efficacy and cost-effectiveness considerations do not support similar routine male screening.2 However, infected men are a significant contributor to female reinfection rates, and cephalosporin-resistant GC strains requiring dual therapy are increasingly identified.3 ,4 As men comprise approximately half of GC infections, and as the majority will remain asymptomatic, the absence of routine screening may be hindering effective female infection intervention and aiding growing antimicrobial resistance.
There is also increasing data concerning both the prevalence of extragenital infections and the potential for pharyngeal-to-genital transmission. Pharyngeal GC (GCoral) infections are more difficult to medically eradicate, may be a reservoir of antimicrobial-resistant infection and like CToral are commonly asymptomatic.5 ,6 Recent studies of heterosexual pharyngeal infection have found that 3.5–6.8% of adolescent female hospital and clinic patients had GCoral; prevalence of CToral/GCoral among Japanese men with urethritis was 2.4%/11.9%, respectively, increasing to 9.1%/25.0% among those with chlamydial or gonococcal urethritis; 23% of individuals with GCoral infections lacked concurrent genital or rectal infections and that pharyngeal-to-genital infection may occur through oral sex.7–10
Emergency departments (EDs) may be feasible screening venues as nearly 6000 existing US hospitals are widely distributed, provide a substantial proportion of screening services and serve a population at known increased risk of CT/GC infection.11 While there exist significant barriers to CT/GC clinical diagnosis in the ED and routine screening is not readily amenable to the acute care environment, effective algorithms identifying those at increased infection risk may be used to allocate resources.12 The purpose of this work was to identify the extent of oropharyngeal CT/GC infection among ED patients and determine if patient-provided information concerning sexual risk factors was predictive of infection status.
Methods
The study was conducted June 2012–March 2013 at the Memorial Medical Center (MMC) ED in Springfield, Illinois. The MMC ED serves approximately 200 patients each day, and although it draws from a wider area, the Springfield population itself is in excess of 116 000 residents (75.6% white patients and 18.5% black patients). Inclusion criteria included patients aged 15–34 presenting between 10:00 and 16:00 with a low-acuity complaint and not otherwise involved with visit-based clinical care. Participants completed informed consent and a sexual history survey, provided both urine sample and oropharyngeal swab for CT/GC testing (regardless of presenting complaint or survey responses) and received a $10 incentive. All samples were analysed using nucleic acid amplification: urines were analysed by the BD Viper XTR System with the Chlamydia trachomatis and Neisseria gonorrhoeae Qx Amplified DNA Assays; and swabs by the Gen-Prob TIGRIS System with the APTIMA Combo 2 assay (validated for this sample type). Infection site was identified by organism (CT or GC) and anatomical site (oral, genital, any/both; eg, CTgen, GCany). Surveys were analysed for associations between patient demographics, sexual history question responses and CT/GC infection status. Associations were determined using χ2 for categorical variables with a two-tailed significance level of 0.05 and binary logistic regression for oral infection (dependent). Missing data were removed before analysis. Age was dichotomised into 15–25 (younger; conforms to current routine female screening age limit in the USA) and 26–34 (older). Statistical analyses were performed with SPSS (V19; Chicago, Illinois, USA). Individuals testing positive for any infection were to be directed to our partnering local health department (LHD) for follow-up and treatment.
Results
Over a 9-month period, 653 individuals were approached and 493 (75.4%) agreed to participate. Participants included 301 women and 192 men; had a mean age of 25.2 years (SD 4.9); were 65.7% white patients and 33.3% black patients; 90.5% were sexually active in the past year and 53.3% were currently married, in a civil union or in a serious relationship exceeding 6 months (table 1). There were 104 female and 56 male non-participants with a mean age of 26.1 years and were 75.6% white patients and 23.8% black patients. Gender, age and racial distributions insignificantly differed between participants and non-participants. Thirty-eight individuals were infected with CT and/or GC at any site (CTany/GCany) for a tested population prevalence of 7.7%. All 38 individuals were successfully treated by the LHD. Table 1 describes infection by anatomical site and significant differences by age, gender and race by any infection. Web-only table 1 lists the sexual history questionnaire variables with their full wording and response rate by gender.
Comparisons of participant demographics, survey responses and infection by age, race and gender
There were 10 individuals infected orally, four with CToral (two each men and women) and another six with GCoral (four men and two women). Those orally infected were no more likely to report “You have sex with:_____” any same sex partners (1 (10.0%) vs 26 (5.5%) of those not orally infected); sexual activity in the past 12 months with any same sex partners (1 (11.1%) vs 38 (9.3%)) or providing any same sex oral sex (1 (20.0%) vs 34 (8.8%%) and none presented with a throat or genitourinary complaint (vs 12 (2.5%) and 38 (7.9%), respectively). Significant univariate associations with oral infection include having a friend with a sexually transmitted disease (4 (44.4%) vs 76 (15.8%); OR=4.25; CI 1.12 to 16.20); reporting anonymous sex in the past year (4 (40.0%) vs 50 (10.4%); OR=5.77; CI 1.58 to 21.15); perception of at least some chance of oral infection (3 (30.0%) thinking chance greater than ‘not at all’ vs 36 (7.5%); OR=5.29; CI 1.31 to 21.28); having two or more partners (any gender) in the past year (8 (80%) vs 182 (39.3%); OR=6.17, CI 1.30 to 29.41); any reported ‘swinging’ or group sex (4 (40.0%) vs 33 (7.0%); OR=8.91; CI 2.40 to 33.14); younger age (9 (3.5%) vs 262 (54.2%); OR=7.59; CI 0.95 to 60.39); not married or in a relationship exceeding 6 months (9 (90.0%) vs 218 (45.4%); OR=10.87, CI 1.36 to 83.33) and race (white patients/black patients/Hispanics orally infected at 5 (1.5%)/4 (2.4%)/1 (20.0%), respectively; p=0.013). Multivariate logistic regression models were examined but not found to be informative or significantly predictive.
Conclusions
Our 9-month pilot project recruited 493 patients aged 15–34. Our tested population prevalence was 7.7% (CI 5.3% to 10.1%), increasing to 11.8% (CI 7.9% to 15.7%) for those aged 15–25, and men were as likely to be infected as women. Furthermore, 2.0% (CI 0.8% to 3.2%) of the population had CT or GC oral infection, increasing to 3.3% (CI 1.1% to 5.5%) of those aged 15–25. Of particular interest are the five individuals (13.2% (CI 2.4 to 24.0%) of all infected) who would have been incorrectly diagnosed and treated if only screened by the routine urine test. One individual had CToral only, but three were GCoral only and one was CTgen/GCoral. Important in light of growing resistance to antibiotics, but also in infection spread and the increased risk of complicated clinical scenarios, these four GCoral-infected individuals represent 25% (CI 3.8% to 46.2%) of all those infected with GC and would have been undiagnosed with standard urine-based screening. Furthermore, focusing screening to only those men reporting having sex with men would have identified only two of the 11 men infected with GC (and only one of the GCoral infections). Although the ED serves an increased risk population, there is often limited space and lack of clinician time for STI screening. Multiple patient factors associated with infection were identified and may be of potential future use for screening algorithm development, but the scope of our pilot study limits their predictive potential. Further work needs to be done to refine screening criteria such that resources are conserved but acceptable levels of those infected are identified.
There are limitations to this study. Although the prevalence of the tested population was high, the total numbers of those infected are relatively low for association analyses, especially at the site-specific level. Second, patients were drawn from a single Midwestern urban hospital, perhaps limiting generalisation to other areas. There is also potential for false-positive results when testing a population of relatively low prevalence, and this may explain the GCoral result from one individual denying sexual activity in the past year. Finally, there is a risk of participation bias as our research coordinator could only approach those who were not otherwise occupied with clinicians or complaint-driven activities. Overall, however, this pilot study has found that not only are ED patients at increased risk of infection with CT and GC, but there is substantial oral infection risk as well. Such infections are frequently asymptomatic, difficult to clinically diagnose and would be missed by urine-based screening. Furthermore, commonly used screening criteria such as gender and sexual activity in the past year were not associated with infection. In light of the continuing high prevalence of CT, the increasing prevalence and concern with cephalosporin-resistant GC and the increasingly understood role of men in female re-infection, the data suggest that further work needs to identify the extent of CT and GC oral infections and mechanisms whereby screening may be feasible and effective.
Key messages
-
Though screening sexually active females under age 26 every year is recommended, chlamydia and gonorrhea (CT/GC) remain the two most reported bacterial infections in the USA.
-
Oral infections (pharynx) with CT/GC are frequently asymptomatic, difficult to identify and treat, and may contribute to genital infection.
-
Relatively high proportions of male and female ED patients are orally and/or genitally infected with CT/GC, and many oral GC infections would not be identified by urine-based screening.
-
Screening of males and extragenital sites may be required to effectively intervene in the continuing prevalence of chlamydia and gonorrhea.
Acknowledgments
Special thanks is given to the Illinois Department of Public Health, Division of Laboratories for the processing of, and genital testing for, the detection of chlamydia and gonorrhoea for this study/project.
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Files in this Data Supplement:
- Data supplement 1 - Online table
Footnotes
-
Handling editor Jackie A Cassell
-
Contributors All three authors fulfil the authorship criteria. WDJ designed and implemented the study, performed the statistical analyses and drafted the manuscript. LLN conducted the field activities, created and maintained the results database and provided manuscript revisions. TC assisted with literature review, designing the study, results interpretation and manuscript drafting and revision.
-
Funding This work was supported by a grant from the Memorial Medical Center Foundation, Springfield, Illinois.
-
Competing interests None.
-
Ethics approval This research was approved by the Springfield Committee for Research Involving Human Subjects.
-
Provenance and peer review Not commissioned; externally peer reviewed.