Sampling technique is important for optimal isolation of pharyngeal gonorrhoea
- 1Melbourne Sexual Health Centre, Alfred Health, Carlton, Victoria, Australia
- 2Melbourne School of Population Health, University of Melbourne, Melbourne, Victoria, Australia
- 3The University of Queensland, St Lucia, Queensland, Australia
- 4Queensland Paediatric Infectious Diseases Laboratory, Royal Children's Hospital, Brisbane Herston, Queensland, Australia
- 5Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, Victoria, Australia
- Correspondence to Dr Marcus Chen, Melbourne Sexual Health Centre, Alfred Health, 580 Swanston Street, Carlton, Vic. 3053, Australia;
- Received 11 February 2013
- Revised 25 April 2013
- Accepted 29 April 2013
- Published Online First 21 May 2013
Background Culture is insensitive for the detection of pharyngeal gonorrhoea but isolation is pivotal to antimicrobial resistance surveillance. The aim of this study was to ascertain whether recommendations provided to clinicians (doctors and nurses) on pharyngeal swabbing technique could improve gonorrhoea detection rates and to determine which aspects of swabbing technique are important for optimal isolation.
Methods This study was undertaken at the Melbourne Sexual Health Centre, Australia. Detection rates among clinicians for pharyngeal gonorrhoea were compared before (June 2006–May 2009) and after (June 2009–June 2012) recommendations on swabbing technique were provided. Associations between detection rates and reported swabbing technique obtained via a clinician questionnaire were examined.
Results The overall yield from testing before and after provision of the recommendations among 28 clinicians was 1.6% (134/8586) and 1.8% (264/15 046) respectively (p=0.17). Significantly higher detection rates were seen following the recommendations among clinicians who reported a change in their swabbing technique in response to the recommendations (2.1% vs 1.5%; p=0.004), swabbing a larger surface area (2.0% vs 1.5%; p=0.02), applying more swab pressure (2.5% vs 1.5%; p<0.001) and a change in the anatomical sites they swabbed (2.2% vs 1.5%; p=0.002). The predominant change in sites swabbed was an increase in swabbing of the oropharynx: from a median of 0% to 80% of the time.
Conclusions More thorough swabbing improves the isolation of pharyngeal gonorrhoea using culture. Clinicians should receive training to ensure swabbing is performed with sufficient pressure and that it covers an adequate area that includes the oropharynx.
Gonorrhoea is one of the most common bacterial sexually transmitted infections worldwide with high prevalence rates reported among many populations of men who have sex with men (MSM).1 ,2 Rectal gonorrhoea has been shown to facilitate the acquisition of HIV.3 Pharyngeal gonorrhoea is believed to be an important reservoir for infection with transmission to the urethra via oral sex.4 Guidelines recommend that MSM should be screened for pharyngeal and rectal gonorrhoea, in addition to other sexually transmitted infections, at least annually, with screening of higher risk men every 3–6 months.5 ,6
Nucleic acid amplification tests (NAAT) and culture are available for the detection of Neisseria gonorrhoeae. NAAT are highly sensitive but false positive results may occur.7 Culture is more specific but less sensitive than NAAT, with studies showing the sensitivity of culture to be as low as 50% compared with NAAT for the detection of pharyngeal gonorrhoea.8 ,9 However, culture permits the determination of antibiotic susceptibility and surveillance of antimicrobial resistance to gonorrhoea, a critical role, given the relentless development of gonococcal resistance to antimicrobials including cephalosporins.10–12 The load of N gonorrhoeae at the pharynx has been shown to be lower than that seen with rectal gonorrhea, with culture from the pharynx missing lower-load infections.8
Pharyngeal gonorrhoea contributes significantly to the epidemiology of gonococcal infection in a number of ways. It is a predominantly asymptomatic infection and therefore will only be detected with screening. Furthermore, it is believed to play a role in the development of antimicrobial resistance via the exchange of genetic material between N. gonorrhoeae and commensal Neisseria species within the pharynx.13 An additional concern is that many of the antimicrobials used to treat pharyngeal gonorrhoea have reduced penetration of the pharynx.14
In a previous study, we found that clinicians who reported inducing a gag reflex more frequently had higher detection rates for pharyngeal gonorrhoea.15 Our hypothesis was that because of the insensitivity of culture, thorough swabbing was required for optimal isolation of pharyngeal gonorrhoea. The aim of the current study was to ascertain whether recommendations on swabbing technique could lead to improvements in pharyngeal gonorrhoea detection rates and to determine which aspects of swabbing technique are important for optimising yield.
This study was undertaken at the Melbourne Sexual Health Centre, the major public sexually transmitted infection clinic in Victoria, Australia. The centre operated a predominantly walk-in, nurse-triaged service where patients were seen by clinicians—doctors and nurses—in the order in which patients presented to the clinic.
In May 2009, following the findings of our initial study, we provided written feedback to individual clinicians at the clinic with a summary of the previous study's findings, the clinician's individual detection rate for pharyngeal gonorrhoea and how this compared with the overall detection rate among all clinicians. Clinicians were advised regarding the likely importance of pharyngeal swabbing technique for optimal gonorrhoea isolation, with adequate coverage of the oropharynx recommended. Screening of MSM at the clinic followed national guidelines,6 with MSM routinely offered testing for pharyngeal gonorrhoea during clinic visits for screening. Pharyngeal specimens were collected using cotton-tipped swabs, which were immediately applied onto modified Thayer Martin medium and forwarded to the onsite laboratory (Microbiology Diagnostic Unit Public Health Laboratory, University of Melbourne) for culture.
In this study, we compared detection rates for pharyngeal gonorrhoea for individual clinicians who were practising at the centre before (June 2006—May 2009) and after (June 2009—June 2012) the above recommendations were provided. Associations between detection rates and reported changes in swabbing technique were examined.
Data on testing for pharyngeal gonorrhoea were obtained for MSM from the clinic's computer database. Only data from MSM were included in the study, defined as males who had at least one male sexual partner in the previous 12 months.
Clinicians were included in this study if they were employed at the clinic over the entire 6-year study period and if they took part in the previous study questionnaire. Clinicians completed a questionnaire to determine whether they had changed their technique for obtaining pharyngeal swabs for gonorrhoea in response to the recommendations. For this analysis, clinicians were grouped into those who reported changing their technique after May 2009 and those who did not. Clinicians who reported a change in technique were asked to indicate whether their technique had changed in relation to the surface area sampled, the swab pressure applied, the frequency with which the gag reflex was induced, the time taken to perform the swab and the anatomical sites swabbed. In our previous study, we asked clinicians how frequently they swabbed the following anatomical sites: the tonsils, the oropharynx and the tonsils plus the oropharynx. We repeated these questions so that before and after responses could be compared directly.
Analyses were performed using SPSS V.21.0. The χ2 test was used to compare categorical variables. Ethical approval for this study was granted by the Alfred Hospital Research Ethics Committee: project approval number 163/12.
Over the 6-year study period, there were 106 364 visits by MSM to the clinic, from whom 32 855 pharyngeal specimens were obtained. The prevalence of pharyngeal gonorrhoea seen among MSM attending the clinic did not change significantly over the study period (figure 1, p for trend=0.5). Over the same period, there were 50 813 clinic visits by MSM seen by the 28 doctors and nurses who had continuous employment over the study period, with 23 632 pharyngeal specimens obtained by these clinicians. The detection rates for pharyngeal gonorrhoea for individual clinicians before and after the recommendations were made are shown in table 1. The overall prevalence of pharyngeal gonorrhoea was 1.6% (134/8586) and 1.8% (264/15046) in the before and after periods respectively (p=0.34). Two clinicians had significant differences in their before and after detection rates—both had significantly lower detection rates in the after period. There was no significant difference in the overall detection rate for all clinicians combined before (1.6%) and after (1.8%) feedback was provided (p=0.17). Nor was there any significant correlation between detection rates by individual clinicians when these were compared for the before and after periods (r=0.19, p=0.33).
Half (n=14) the clinicians in the study indicated that they had changed their pharyngeal swabbing technique since the recommendations were provided: all 14 reported swabbing a larger surface area, eight applied more pressure, 14 induced gag reflexes more frequently, 12 spent more time swabbing and 11 had changed the anatomical sites they swabbed.
Detection rates among clinicians, grouped according to whether they had changed specific aspects of their swabbing technique, are shown in table 2. Compared with clinicians who did not change their swabbing technique, significantly higher detection rates were seen among clinicians who had changed their swabbing technique in response to the recommendations (2.1% vs 1.5%; p=0.004), swabbed a larger surface area (2.0% vs 1.5%; p=0.02), applied more swab pressure (2.5% vs 1.5%; p<0.001) and changed the anatomical sites which they swabbed (2.2% vs 1.5%; p=0.002).
While none of the increases in detection rates for individual clinicians was statistically significant, clinicians with higher detection rates in the period following the recommendations were, overall, more likely to have changed their swabbing technique. Sixty-nine percent (9/13) of the clinicians who had a subsequent increase in detection rates reported changing their technique, while only 33% (5/15) of clinicians whose detection rates had not increased reported a change in technique (p=0.058).
The median frequencies with which specific anatomical sites were swabbed before and after the recommendations are shown in table 3. Prior to the recommendations, clinicians predominantly obtained samples from the tonsils, with few swabbing the oropharynx. Following the recommendations, the major change that occurred was an increase in swabbing of the oropharynx: the median frequency with which the oropharynx was swabbed increased from 0% to 80% of the time.
The median frequency with which clinicians reported inducing a gag reflex increased from 50% before the recommendations to 80% after. The median change in this frequency across clinicians as a group was 15%. Clinicians who had a change in the frequency with which they induced a gag reflex above the median (>15%) had significantly higher detection rates than clinicians who did not (p<0.001).
The findings of this study indicate that the technique used to obtain specimens from the pharynx for gonorrhoea isolation is important for optimal detection. We found that providing clinicians with individual feedback regarding their detection rates and recommendations on the importance of thorough specimen collection positively influenced clinical practice among half of these clinicians, which led to an improvement in detection rates among those who changed their technique. Overall, clinicians who changed their sampling technique in response to the recommendations had significantly higher detection rates for gonorrhoea than those who did not. Specifically, clinicians who reported swabbing a larger surface area, applying more swab pressure and expanding the anatomical sites they swabbed had significantly higher detection rates than those clinicians who did not.
This study has several limitations. First, this was a retrospective observational study where factors other than clinician sampling may have influenced gonorrhoea isolation rates. Clinicians were asked to report on their pharyngeal swabbing technique and whether their technique had changed over time. Therefore, the responses may have been subject to recall bias. However, several different measures of sampling technique were found to be significantly associated with improved gonorrhoea isolation and all are consistent with the need to obtain more specimen material. Second, the detection rates for pharyngeal gonorrhoea are dependent on the underlying prevalence of pharyngeal gonorrhea, which may have varied over time. However, the overall prevalence of pharyngeal gonorrhoea for the clinic population of MSM did not change significantly over the study period. Third, while the study was well powered to detect differences in detection rates when clinicians were combined, power was more limited when comparing detection rates for individual clinicians before and after the recommendations.
We believe that because of the poor sensitivity of culture for pharyngeal gonorrhoea, perhaps compounded by the relatively low bacterial load of gonorrhoea at the pharynx, more thorough swabbing is required to optimise the detection of pharyngeal gonorrhoea using culture. We postulate that the use of more pressure and coverage of a larger area that includes the tonsils and the oropharynx increases the amount of organisms captured on the swab. The association between more frequent induction of the gag reflex and higher detection rates in this and our previous study15 most probably reflects a more vigorous swabbing technique.
Although both NAAT and culture are used for the diagnosis of pharyngeal gonorrhoea, the use of culture has diminished over time, with a shift towards greater use of NAAT. Some authorities now recommend the preferential use of NAAT for screening for pharyngeal and rectal gonorrhoea.5 ,16 As NAAT are more sensitive for the detection of pharyngeal gonorrhoea,8 ,16 it is possible that swabbing technique may be less critical for the detection of pharyngeal gonorrhoea using NAAT, though this needs to be evaluated. Despite its poorer sensitivity, culture will remain important for the determination of antimicrobial susceptibility and surveillance of antimicrobial resistance, whether it is applied routinely, as in our clinic, or only in circumstances of suspected treatment failure.10–12 Clinicians who screen for pharyngeal gonorrhoea should receive training to ensure swabbing is performed with sufficient pressure and that it covers an adequate area that includes the oropharynx. This should be done in a way that minimises patient discomfort.
Clinicians who changed their pharyngeal swabbing technique by applying more swab pressure and by swabbing a larger surface area including the oropharynx had significantly higher Neisseria gonorrhoeae isolation rates.
Feedback and recommendations to clinicians on their screening practice had a positive influence on half of clinicians, which led to a measurable improvement in outcomes among those who responded to the recommendations.
Training and education on the specific technique required to obtain an optimal sample for pharyngeal gonorrhoea are required.
We thank Afrizal Afrizal and Jun Kit Sze for providing data on testing, Huachun Zou and Lenka for statistical assistance, and doctors and nurses from the Melbourne Sexual Health Centre for participating in the questionnaire.
Handling editor Jackie A Cassell
Contributors MM and VR were responsible for data collection and analysis. All authors were involved in the study design, interpretation of data and writing of the manuscript.
Competing interests None.
Ethics approval Alfred Hospital Research Ethics Committee.
Provenance and peer review Not commissioned; externally peer reviewed.