Objective Previous studies have quantified bacterial loads of Neisseria gonorrhoeae in the pharynx and rectum of men but not the urethra. We quantified the bacterial load of N. gonorrhoeae in men with symptomatic and asymptomatic urethral gonorrhoea infections.
Methods Consecutive men diagnosed with urethral gonorrhoea by Aptima Combo 2 testing of urine at the Melbourne Sexual Health Centre between March and July 2016 were eligible for the study: symptomatic men with purulent urethral discharge and asymptomatic men with no urethral symptoms. The gonococcal bacterial load in both groups was measured by urethral swab using a standardised collection method and real-time quantitative PCR targeting the opa gene.
Results Twenty men were recruited into the study: 16 had purulent urethral discharge and 4 had asymptomatic urethral gonorrhoea. The median gonococcal bacterial load was significantly higher among symptomatic men (3.7×106 copies per swab, IQR 2.5×106–4.7×106) compared with asymptomatic men (2.0×105 copies per swab, IQR 2.7×104–4.5×105) (p=0.002).
Conclusions Gonococcal loads in men with urethral discharge were higher than loads seen with asymptomatic urethral gonorrhoea and loads seen in asymptomatic pharyngeal and rectal gonorrhoea infections in previous studies.
- NEISSERIA GONORRHOEA
- BACTERIAL INFECTION
- MODES OF TRANSMISSION
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
Neisseria gonorrhoeae infection of the male urethra typically presents with purulent urethral discharge. Some clinic-based studies estimate urethral gonorrhoea to be asymptomatic in approximately 10% of men.1 If the bacterial load is a determinant of the transmissibility of gonorrhoea,2 then comparative bacterial loads from different anatomical sites of infection could provide an indication of the relative transmissibility of gonorrhoea from different sites. Previous studies have measured bacterial loads of N. gonorrhoeae in the pharynx and rectum of men but not the urethra.3 ,4
The aim of this exploratory study was to quantify bacterial loads of N. gonorrhoeae in men with symptomatic and asymptomatic urethral gonorrhoea. Our first hypothesis was that bacterial load in the urethra would be higher in men with urethral discharge compared with asymptomatic men. Our second hypothesis was that bacterial load in the urethra of men with urethral discharge would be higher than loads found in asymptomatic pharyngeal and rectal infections in previous studies.
This was a prospective cross-sectional study of men diagnosed with urethral gonorrhoea at the Melbourne Sexual Health Centre between March and July 2016. Men with urethral gonorrhoea were categorised as symptomatic or asymptomatic. Symptomatic men reported symptoms of urethritis and had urethral discharge on examination. Asymptomatic men reported no urethral symptoms and had no discharge on examination. Men receiving any antibiotics within 7 days were excluded. Men reporting sex with men (MSM), as well as men reporting sex with women, were eligible.
Data were obtained on time between last sexual contact of any type and duration of urethral symptoms, if present. All men were tested for gonorrhoea and chlamydia using first pass urine and the Aptima Combo 2 (AC2) assay for screening, and the Aptima gonococcus (GC) test for confirmation of all positive samples on the Hologic Panther platform. Culture for gonorrhoea was not performed.
A urethral swab was collected from each man, whether symptomatic or asymptomatic, to provide an estimate of bacterial load prior to treatment. This was taken at the initial clinic visit for symptomatic men, and at a second visit in asymptomatic men, following recall for treatment. Asymptomatic men were included if a repeat urine AC2 test was positive from the second visit, which occurred at a median of 3.5 days (IQR 2.5–7.5) after the first visit. The repeat AC2 was to ensure the first positive result was a true positive and current infection. A Copan FLOQSwab (Copan Italia SpA, Brecia, Italy), a sterile neonatal nasopharyngeal swab in a dry tube, was inserted into the urethra without rotation such that the entire 9 mm length of the cotton tip was inserted fully into the distal urethra. All swabs were taken by DP or JJO, who were trained in the procedure to ensure consistency in sampling between men.
Swabs were rotated in 400 µL of phosphate-buffered solution (PBS) in a 1.5 mL phial, and the swab discarded. Released cells in PBS were stored at −80°C before analysis by real-time quantitative PCR (qPCR). During analysis, 200 µL of PBS was extracted using a MagNApure 96 (Roche Diagnostic, Indianapolis, Indiana, USA) using the DNA and viral nucleic acid small volume kit and eluted in 100 µL of elution buffer. To assess sample adequacy, a 260 bp region of the human β-globin gene was amplified by qPCR using previously described primers5 ,6 and the following taqman probe: 6FAM-ACACAACTGTGTTCACTAGC-TAM (PCO3); the results were then adjusted to account for haploid copies of the β-globin gene. N. gonorrhoeae copies were amplified by qPCR from 5 µL of eluate, with bacterial load calculated by comparing the crossing points with a standard curve, constructed with known concentrations of gonorrhoeal cells, as described previously.3 ,7 ,8 This method of plotting against known cell numbers accounts for both the 11 copies of the opa gene found within N. gonorrhoeae and the polyploid nature of N. gonorrhoeae. 9 ,10 The final concentration was calculated by multiplying the DNA load by the dilution factor and reported as copies per swab.
Median values and IQRs for gonococcal load and β-globin were calculated. The correlation between load and β-globin was calculated using the Kendall rank correlation coefficient. Bacterial loads in symptomatic and asymptomatic men were compared using the Mann-Whitney U test. The study was approved by the Alfred Health Human Research Ethics Committee (97/16).
Of the 20 men eligible for the study, 16 had symptomatic and 4 had asymptomatic urethral gonorrhoea. Excluded from the study was one man with gonorrhoea who had received recent antibiotics. The characteristics of men and gonococcal bacterial loads are shown in table 1. Median age was 32 years. Sixteen men reported sex with other men. The median interval between last sexual contact and measurement of load was 6.5 days (IQR 5.0–10.0). Among symptomatic men, the median duration of symptoms prior to measurement of load was 2.5 days (IQR 1.5–4.0).
β-globin was detected in all urethral swab specimens. The median β-globin level was significantly higher among symptomatic men (2.1×107 copies per swab; IQR 5.7×106–3.0×107) compared with asymptomatic men (5.1×105 copies per swab; IQR 5.0×104–9.2×105) (p=0.002). Furthermore, there was a significant correlation between quantity of β-globin and gonococcal load (τ=0.43, p=0.01) (figure 1).
Median gonococcal load among the 16 men with discharge on examination (3.7×106 copies per swab; IQR 2.5×106–4.7×106) was significantly higher than median load for the four asymptomatic men (2.0×105 copies per swab; IQR 2.7×104–4.5×105) (p=0.002)(figure 2). There was no association between load and time since last sexual contact or duration of symptoms (data not shown).
To our knowledge, this is the first study to quantify the load of N. gonorrhoeae in men with urethral gonorrhoea using qPCR. The results demonstrate that gonococcal load in the urethra is higher in men with urethral discharge on examination compared with asymptomatic men with no discharge: median 3.7×106 vs 2.0×105 copies per swab respectively. The results also suggest that gonococcal loads in men with urethral discharge are higher than loads seen in asymptomatic pharyngeal and rectal gonorrhoea infections in MSM in previous studies.3 ,4 Summarising previous inoculation studies of male urethral gonococcal infections using culture, Hobbs et al postulated the probability of infection may increase with the inoculum dose as measured by colony-forming units.2 However, data supporting this are limited. Experimental inoculation studies of the male urethra11 have quantified the dose of N. gonorrhoeae needed to produce infection in 50% of experimental subjects. However, we do not know how these doses relate to the bacterial loads obtained using our qPCR assay. Future studies of quantitative culture in parallel with load assay may be helpful to quantify this relationship.
In a previous study of MSM attending our clinic, the median gonococcal load in pharyngeal infections was 2100 copies per swab, which was significantly lower than the median load of 18 960 copies per swab for rectal infections. Furthermore, the median load in MSM with symptomatic gonococcal proctitis was 278 800 copies per swab, which was significantly higher than the median of 13 980 copies per swab among men with asymptomatic rectal gonorrhoea.3 In a more recent study of MSM attending our clinic, the median gonococcal load in PCR-positive pharyngeal infections was 248 512 copies per swab and 170 493 copies per swab in culture-positive and culture-negative men, respectively.4 The substantially higher pharyngeal loads in the second study likely reflect differences in study design and sampling. In the first study by Bissessor et al, swabs were taken by clinicians during routine clinical care primarily from the tonsils only.3 ,12 ,13 In the second study by Chow et al, sampling was undertaken by clinicians trained to use a standardised method for collection from the tonsils and posterior oropharynx, and men had culture-positive pharyngeal gonorrhoea, which may have biased towards higher loads.3 ,4 ,14
The present study used the same method for measurement of gonococcal load as the studies by Bissessor and Chow: real-time qPCR targeting the opa gene. Direct comparison of loads between studies should be undertaken with caution, given potential variability in sampling between studies and differences in anatomical sites infected. We found a correlation between β-globin and gonococcal load. Although we undertook β-globin quantification to assess sampling adequacy, we believe this correlation may also be due to increased inflammatory cells present in men with discharge, as sampling was performed by the researchers using a consistent method. β-globin levels solely reflecting adequacy of sampling should be broadly comparable across samples irrespective of whether discharge was present or not. However, β-globin levels were significantly higher in symptomatic men, with asymptomatic men having the lowest β-globin loads (figure 1).
The high gonococcal loads seen among men with urethral discharge—which were on average 15-fold higher than loads in pharyngeal infections in the study by Chow—suggests the purulent discharge of urethral gonorrhoea may be relatively contagious. Men with symptoms suggestive of possible urethral gonorrhoea should thus be encouraged to present early for treatment to prevent further transmission. While men with asymptomatic urethral gonorrhoea had lower loads, these were comparable with those seen in pharyngeal infection.4 This suggests asymptomatic gonococcal urethral infection may also be potentially infectious and provides support for guidelines that recommend screening asymptomatic MSM for urethral gonorrhoea.
Gonococcal bacterial loads were higher among men with purulent urethral discharge compared with men with asymptomatic urethral gonorrhoea.
Gonococcal loads in men with purulent discharge were higher than loads in asymptomatic pharyngeal and rectal gonorrhoea infections from previous studies.
The purulent discharge characteristically seen with urethral gonorrhoea may be relatively contagious compared with asymptomatic gonococcal infections at extragenital sites.
The authors thank the patients who participated and the staff at Melbourne Sexual Health Centre who helped recruit patients to the study.
DP and JJO are equal first authors.
Handling editor Jackie A Cassell
Contributors The study concept was conceived by MC and designed by DP, JJO, ST and MC. DP and JJO undertook sampling and data collection. SG, ST and SP oversaw laboratory testing and analysis. All authors contributed to interpretation of the data, writing of the manuscript and approval of the final version.
Funding JJO, EPFC and TRHR are supported by the Australian National Health and Medical Research Council Early Career Fellowships (No. 1104781, 1091226, 1091536 respectively).
Competing interests None declared.
Ethics approval Alfred Health Human Research Ethics Committee, ethics approval number is (97/16). All patients provided written consent to join the study.
Provenance and peer review Not commissioned; externally peer reviewed.