Diagnosis,management and prevalence estimation of gonorrhoea: influences of Aptima Combo 2 assay with alternative target confirmation

Abstract

Case-notes and laboratory data were retrospectively reviewed for influences of dual testing by Aptima Combo 2 (AC2) for Chlamydia trachomatis (CT) and Neisseria gonorrhoeae (NG) on the diagnosis, management and prevalence estimation of gonorrhoea in the genitourinary (GU) medicine clinic and community. NG positives by AC2 were confirmed by Aptima Gonococcus assay. Unconfirmed positives were rare. Our study showed that in the GU medicine clinic, AC2 detected about 20% extra cases of NG beyond culture. For best standard of care, NG culture and microscopy are still required in some patients to ensure that treatment is rapid and appropriate. Compared to self-referral at the GU medicine clinic, community tests made a substantial contribution to the overall number of NG cases found (40 community versus 35 Macclesfield GU medicine clinic). The ratio of female to male NG cases found was significantly higher (P = 0.002) in the community (13 M, 27 F) than at the GU medicine clinic (25 M, 10 F). In the community, over 60% of NG infections occurred in chlamydia-negative patients. The overall prevalence of NG in the GU medicine clinic was 1.3%, the true prevalence being much lower at 0.9% on primary test. Prevalence in the community was 0.4%. Data from dual testing in the community can clarify NG prevalence beyond the existing KC60 (sexually transmitted infections) reports.

Keywords

Aptima Combo 2 nucleic acid amplification tests confirmation dual testing gonorrhoea culture management prevalence

INTRODUCTION

Culture remains the current ‘gold standard’ for the diagnosis of Neisseria gonorrhoeae (NG). However, as NG is a fastidious organism, the sensitivity of detection may be affected by suboptimal collection, transport and incubation. Nucleic acid amplification tests (NAATs) are being increasingly introduced for the diagnosis of NG and have considerable advantages over culture, including high sensitivity across a range of specimen types and use under varying settings and conditions.^1,2 However, there are concerns about specificity of NAATs in low-prevalence NG populations and its associated false positivity (FP), impacting on positive predictive values.^3–5 FP can occur with some NAATs as their target may cross-react with different species of Neisseria.^2,6 The evidence may be less for Aptima Combo 2 (AC2, Gen-Probe, San Diego, CA, USA), but to date this type of FP has not been observed with AC2 or Aptima Gonococcus (AGC) assays.^1,7

AC2, a dual assay for detection of both Chlamydia trachomatis (CT) and NG, has been used in our GU medicine clinic and in surrounding community settings for diagnosis and screening since early 2006. Samples initially reactive for NG by AC2 are confirmed by re-testing the residual sample against an alternative NG-specific RNA target in the AGC assay.

This retrospective study was to review influences of dual testing for CT and NG by AC2 on the diagnosis, management and prevalence estimation of gonorrhoea infections in both the GU medicine and community settings.

METHODS

Study period

The study period was from 8 August 2006 to 4 April 2008.

Study population

We reviewed laboratory data on AC2 testing for CT and NG and case-notes of all patients with a positive result for NG by culture or by AC2 at both the Macclesfield GU medicine clinic (3589 AC2 samples from 1930 female patients and 2470 AC2 samples from 1867 male patients) and the corresponding East and Central Cheshire community served by the ‘Team Chlamydia’ office of the National Chlamydia Screening Programme, which organizes all ‘non-GU medicine’ screening and diagnostic AC2 dual testing in that area (7934 female and 1549 male samples).

Additionally, we refer to 107 NG cases reported by a second GU medicine clinic (Leighton), which serves the same community (M Grobicki, personal communication).

Samples tested

Culture was performed on urethral swabs from male patients and on both urethral and endocervical swabs from female patients. Microscopy was performed on urethral and endocervical swabs from symptomatic patients. Where history was suggestive, rectal and throat swabs were also tested.

Urethral swabs, vulvovaginal, endocervical swabs and, where history was suggestive, rectal and throat swabs were collected in appropriate AC2 Swab Collection Kits. First-void urine was collected in AC2 Urine Collection Kits after at least one hour of bladder retention.

Laboratory methods

The AC2 assay for NG was performed according to the manufacturer's instruction. Samples initially reactive for NG by AC2 were confirmed by re-testing residual sample against an alternative NG-specific RNA target in the AGC assay.

Swabs for NG culture were transported in Amies charcoal transport medium to the local microbiology laboratory within 2–3 hours of sampling and were plated on selective culture medium (Oxoid, Perth, Scotland) and incubated at 37°C with 5% CO₂ for 48 hours before examination. Suspected colonies were confirmed by Gram stain, biochemical and immunological tests.

Data items/analysis

Age, gender, sexuality, referral pattern, symptoms, concomitant infection and contact tracing were extracted retrospectively from patient case-notes and analysed using SPSS version 16. Laboratory data analysis was performed by using EpiInfo version 6 with statistics by chi-square with Yates correction. Prevalence calculation in the community setting was based on sample positivity as the influence of repeat testing is considered negligible.⁸

RESULTS

Diagnosis and management

Specificity of AC2 assay

Of the total of 15,542 tests performed at both community and GU medicine clinics, only one was positive for NG by AC2 but unconfirmed by AGC assay.

NG cases found at the GU medicine clinic

During the study period, Leighton GU medicine clinic serving this same community reported 107 cases of NG (41 F, 66 M).

At the Macclesfield GU medicine clinic, a total of 52 (20 F, 32 M) diagnoses of NG were made by either AC2 or culture. Forty-seven patients (Figure 1) were AC2 positive and had both AC2 tests and culture (5 cases excluded – 3 had AC2 only and 2 had cultures only).

Figure 1

Findings in 47 Neisseria gonorrhoeae (NG) positive cases tested at the genitourinary medicine clinic by both Aptima Combo 2 (AC2) and culture

Table 1 shows the patterns of testing and results for the four (25%) female patients and six (19%) male patients who would have been missed if culture alone had been performed. In four of these male cases, the AC2 throat swab result was key to the diagnosis.

Table 1

Pattern of samples tested and results for 10 patients with AC2 positive but culture negative findings

Sex	Vaginal swab	Cervix swab	FVU	Urethra swab	Throat swab	Rectal swab	Partner(s) with NG
F		A+ C−		A+ C−			2
F	C−	C−	A+	C−
F	C−	A− C−		A+ C−
F	A+ C−
M				A− C−	A+ C−	A−
M				A+ C−
M			A+	A+ C−	A+ C−	A− C−
M			A−		A+ C−	A− C−	1
M			A−		A+ C−		1
M			A−		A+ C−

A = Aptima Combo 2 result (positives confirmed by Aptima Gonococcus assay); C = NG culture result; FVU = first-void urine; NG = Neisseria gonorrhoeae

NG cases found in the community

In the community, 40 cases of NG (27 F, 13 M) were diagnosed by AC2 (38 records available for a detailed review).

Characteristics of NG cases found at the GU medicine clinic and in the community

Table 2 compares characteristics of the NG cases found at the Macclesfield GU medicine clinic with those diagnosed in the community. Female patients were significantly younger compared with male patients at both the GU medicine clinic (P = 0.001) and the community (P = 0.02); age profile appeared similar in GU medicine clinics and community.

Table 2

Characteristics of NG-positive cases at the GU medicine clinic versus Community

	Macclesfield GU medicine clinic n = 52	Community n = 40
Age
Range	16–66	15–61
Mean age for male patients*	33.6	30.2
Mean age for female patients*	22.8	21.6
Gender
Male	32 (62%)	13 (32.5%)
Heterosexual	20/32 (63%)	13/13 (100%)
Homosexual	11/32 (34%)
Bisexual	1/32 (3%)
Female	20 (38%)	27 (67.5%)
Referred by
Self	35 (25 M, 10 F) = 67%	40 (13 M, 27 F)
Team Chlamydia	6 (1 M, 5 F) = 12%
General practitioner	1 (1 M) = 2%
Contact slips	10 (5 M, 5 F) = 19%
Concomitant chlamydia infection	M = 12/32 (37.5%)	M = 5/13 (38.5%)
Concomitant chlamydia infection	F = 11/20 (55%)	F = 10/27 (37%)
Symptoms		(38 records reviewed)
	M = 24/32 (75%)	M = 9/13 (69%)
	F = 10/20 (50%)	F = 10/25 (40%)
Management strategies	Total no. of contacts of 52 NG cases = 100	Fast track to GU medicine clinic as first option^†
		Contact tracing pursued by respective GU medicine clinics
	Total no. of contacts traced and treated = 43 (43%)	Macclesfield GU medicine clinic – 6
	Total no. of contacts found NG positive = 7/43 (16.3%)	Leighton GU medicine clinic – 17
		Prison GU medicine clinic – 2
		Withington GUmedicine clinic – 1
		General practice – 3
		Private clinic – 1

NG = Neisseria gonorrhoeae; GU = genitourinary; M = male patients; F = female patients

* = Female patients were significantly younger compared with male patients at both GU medicine clinic (P = 0.001) and community (P = 0.02) by independent sample t-test

^† = Of the 38 cases reviewed, referral location was noted in 30 and therapy detail were available for 25

When considering self-referral (primary diagnosis), the testing in the community made a substantial contribution to the overall number of NG cases found (40 community versus 35 Macclesfield GU medicine clinic). The ratio of female to male NG cases found was significantly higher (P = 0.002) in the community (13 M, 27 F) than at the GU medicine clinic (25 M, 10 F).

For all male patients and for female patients in the community, over 60% of NG infections occurred in the absence of concomitant chlamydia infection. Community NG cases were found more among the chlamydia-negative than among the chlamydia-positive sectors (female patients 17 versus 10, male patients 8 versus 5). Although the overall effect was significant (P = 0.04), the results did not reach significance for female and male patients separately.

Asymptomatic NG infection was more apparent in female patients tested in the community than in the GU medicine clinic, but this did not reach significance.

All patients at the GU medicine department at the study site were treated with third-generation cephalosporins as per national British Association of Sexual Health and HIV guidelines, NG at the genital site being treated with 400 mg of oral cefixime and NG at the pharynx being treated with ceftriaxone 250 mg intramuscularly stat. No resistance to cephalosporins on antibiotic sensitivity in this study was reported; however, resistance to ciprofloxacin, penicillin and tetracycline was 16%, 18% and 6%, respectively.

Of the NG cases found in the community and referred to the GU medicine clinic, 21/25 had oral cephalosporins, one ceftriaxone, one ciprofloxacin and two doxycycline.

Prevalence

For the comparison of prevalence between GU medicine clinic and community, only positive results by AC2 (50 at the GU medicine clinic and 40 in the community) were used (Table 3). A subset of GU medicine clinic figures shows primary tests to better reflect prevalence through exclusion of the high positive rate in case contacts and known NG positives referred from the general practitioner (GP) and Team Chlamydia.

Table 3

Gonorrhoea prevalence by age group and sex in GU medicine clinic settings versus Community

		Male patients No. positive/No. tested (%)		Female patients No. positive/No. tested (%)
Setting	Population	(15–24 years)	(Over 24 years)	(15–24 years)	(Over 24 years)
GU medicine clinic	All tests	9/807 (1.1)	23/1060 (2.2)	10/1053 (0.95)	8/877 (0.91) c*
GU medicine clinic	Primary tests	7/795 (0.88)	18/1030 (1.7)	6/1029 (0.58)	4/853 (0.46)
Community, with subpopulations where information is available	All tests	9/1330 (0.67)	4/219 (1.8)	23/5070 (0.45)	4/2864 (0.14) c*
	FP	0/118 (0)	0/25 (0)	3/1171 (0.26) b*	0/603 (0)
	GP	5/204 (2.5) a*	3/125 (2.4)	8/1653 (0.48)	2/1614 (0.12)
	Prison	NA	NA	5/339 (1.5) b*	2/467 (0.42)
	Youth clinic	2/778 (0.25) a*	0/33 (0)	7/1661 (0.42)	0/121 (0)

FP = family planning; GP = general practice; GU = genitourinary; NA = not applicable; No. = number

a*, P = 0.004; b*, P = 0.02; c*, P = 0.002

The only statistically significant difference in prevalence between GU medicine clinic and community was for female patients aged over 24, and even this difference lost significance when only primary tests were considered.

In the community, NG positivity in younger male patients was significantly higher in general practice than at youth clinics. For female patients, NG cases were found across a wider distribution of services but with somewhat higher prevalence in younger female patients in prison.

DISCUSSION

Not all NAAT platforms perform equally on all sample sites. On female urine, Roche's Cobas Amplicor PCR assay (Roche Diagnostics Ltd., Burgess Hill, West Sussex, UK) and Becton Dickinson's strand displacement assay (SDA. BD Diagnostics – Diagnostic Systems, Oxford, UK) perform less well as compared with AC2.^9,10 Cook et al. ¹⁰ conclude that the sensitivity of PCR in female subjects is too low to recommend its routine testing for NG in urine specimens.

Barlow¹¹ emphasizes that a different NAAT with a different target must be used to confirm a positive result. The Centers for Disease Control and Prevention (CDC) recommends confirmation of positive NAAT results in low-prevalence populations.¹² Since few laboratories can have access to a second NAAT platform, Boyadhzhyam et al. ¹³ looked at whether the Aptima CT (ACT) and AGC assay, which targets rRNA sequences different from AC2, could be used to confirm AC2 results. They found 100% correlation between ACT–AGC and AC2. Since AC2, ACT and AGC use identical specimen storage buffers and conditions, similar procedures and equipment, confirmatory testing with ACT–AGC following AC2 is a very convenient and easy approach to implement in a clinic laboratory setting. Golden et al. ¹⁴ found that the routine confirmation of NG-positive specimens using the AC2 assay was not necessary in a low (0.5%)-prevalence population. Moncada et al. ⁷ evaluated three of the CDC approaches for confirming NG-positive NAAT results and concluded that confirmatory testing was not warranted for genital specimens. They also found that AC2 and AGC assays were better at confirmation than SDA and PCR.

Our report here is only on the Aptima platform but, as in other studies, has shown the AC2 assay for NG to be highly sensitive and specific.^9,15,16 There were no culture-positive but AC2-negative results in any of our patients. Of the total of 15,542 tests performed both in the community and at the GU medicine clinic, only one was positive for NG by AC2 but unconfirmed by AGC. Until more evidence accumulates, it is advisable to follow a general guideline that confirmation should be the norm. However, our results and those of others suggest that, with extra evidence on AC2 in the future, confidence may grow that confirmation is unnecessary – this may depend on examination of the stability in NG of the two sequences targeted by AC2 and AGC assays.

NAATs are being increasingly introduced for the diagnosis of NG and general guidelines are welcome – such as those issued for Scotland⁹ covering platform sensitivity, confirmatory tests for specificity, importance of specimen type and clinical scenario for continuing NG culture.

Dual testing of CT and NG by AC2 at no extra cost has led to increased detection of NG cases both in the community and at the GU medicine clinic. The 40 NG cases found in the community represent a 25% increase over the 159 (52 + 107) diagnosed at both GU medicine clinics in the same period. Twenty-three young female subjects could have gone undiagnosed and untreated for NG within the community if the NAAT performed had been for chlamydia alone. This raises the issue of whether testing for NG should become a routine adjunct in the chlamydia screening population. Since, in our study, the majority of NG cases were found in the chlamydia-negative population, dual testing at the outset would find more NG infections than reflex testing of chlamydia positives.

Within the GU medicine clinic, pick-up of extra cases probably reflects superior sensitivity of AC2; 10 out of 47 cases would have been missed if tested only by culture – throat swabs positive by AC2 but negative by culture being an important contribution to male subjects.

Although not approved by the Food and Drug Administration but because of high specificity of all CT NAAT platforms, non-genital specimens have been validated and are now widely used for testing rectal and pharyngeal swabs.^17,18 However, NAATs on extra genital sites for NG are not currently recommended. The sensitivity of culture for detection of NG at the pharynx is only around 40–50% and that at the rectum is around 50–60%.⁹ Because of the reduced sensitivity of NG culture in the rectum and pharynx, it has been suggested that molecular tests may be more appropriate for the detection of NG at these sites.¹⁹ Commensal Neisseria species at the pharynx cross-react with SDA and PCR assay, but to date this has not been reported for AC2 and AGC assays.^1,7 We have found AC2 to be useful.

An important limitation of NG NAATs is inability to generate antibiotic susceptibility data. Therefore there is the need to maintain culture methods. The aim of gonococcal antimicrobial resistance surveillance is to determine a suitable treatment, which cures a minimum of 95% of infections when given as a single dose, preferably on diagnosis or first presentation.²⁰ Microscopy in symptomatic patients facilitates rapid treatment; however, culture should be taken in GU medicine clinic patients to facilitate appropriate treatment in those who are symptomatic, positive on gram stain, a known contact of gonorrhoea, where treatment has failed and in patients who are NAAT positive and have not had treatment. The recent Gonococcal Resistance to Antimicrobials Surveillance Programme (GRASP annual report 2007)²¹ showed resistance to antimicrobials much higher than found in our GU medicine clinic – it reported 28% resistance to ciprofloxacin, 24% to penicillin, 4% to azithromycin, 60% to tetracycline and for the first time ever found decreased susceptibility to cefixime in two isolates, none were reported to spectinomycin or ceftriaxone.

The NG prevalence reported in different populations in the UK when tested by NAATs varies between 0.5 and 4%.^{9,15,16,22–24} The overall prevalence of NG in our GU medicine clinic is 1.3%; however, the true prevalence is much lower at 0.9% on primary test (after excluding already diagnosed cases referred from the community and those presenting as contacts), whereas prevalence in the corresponding community is 0.4%, which is similar to 0.5% and 0.7% found in other studies.^23,24 Rao et al.,²⁵ however, in a community clinic in South East London found a higher rate of 3.8% in female patients and 5.7% in male patients. In the community, our findings of most male cases at GP may be linked to the fact that most men have symptoms which take them there. In contrast, women who are likely to be asymptomatic were detected at a wider range of services.

The Health Protection Agency in the UK uses the KC60 returns (STI [sexually transmitted infection] diagnoses code) from GU medicine clinics for epidemiological monitoring of STIs. The KC60 may not represent the true burden of disease nationally because it does not include STI diagnoses made outside the GU medicine clinics. Also by including contacts the KC60 will overestimate as shown in our study and supported by others.^26,27 Better ways of collecting data on STIs, so that analysis can be unconfounded, need to be looked into.

CONCLUSIONS

AC2 with AGC confirmation is accurate – both sensitive and specific in detecting NG. It finds more cases than culture. AC2 community testing substantially increases case findings. The prevalence of gonorrhoea in the GU medicine clinic is overestimated by the KC60 coding. Dual testing in the community can clarify NG prevalence beyond the existing KC60.

NG culture and microscopy are still required to provide best standard of care for selected patients.

Footnotes

ACKNOWLEDGEMENTS

We thank Mr Kevin Wright at microbiology department at Macclesfield District General hospital in providing information on culture test, to Janette Nordhall Payne and Sue Shelley of Chlamydia screening office, Congleton, Cheshire East Community Health for data collection. Also to Moira Grobicki for providing data from Leighton GU medicine clinic.

References

Whiley

, Tapsall

, Sloots

. Nucleic acid amplification testing for Neisseria gonorrhoeae – an ongoing challenge. J Mol Diag 2006;8 (1):3–15

Van Dyck

, Leven

, Pattyn

, Detection of Chlamydia trachomatis and Neisseria gonorrhoeae by enzyme immunoassay, culture, and three nucleic acid amplification tests. J Clin Microbiol 2001;39:1751–6

Stanley

, Todd

. Testing for Neisseria gonorrhoea by nucleic amplification testing of chlamydia samples using Roche Cobas Amplicor in a rural area in the north of England does not find more gonorrhoeae in primary care. Sex Transm Infect 2005;81:518

Diemert

, Libman

, Lebel

. Confirmation by 16SrRNA PCR of the Cobas Amplicor CT/NG test for diagnosis of Neisseria gonorrhoeae infection in a low prevalence population. J Clin Microbiol 2002;40:4056–9

Katz

, Effer

, Ohye

, False positive gonorrhoea test results with a nucleic acid amplification test: the impact of low prevalence on positive predictive value. Clin Infect Dis 2004;38:814–9

Palmer

, Mallinson

, Wood

, Evaluation of specificities of five DNA amplification methods for the detection of Neisseria gonorrhoeae . J Clin Microbiol 2003;41 (2):835–7

Moncada

, Donegan

, Schachter

. Evaluation of CDC – recommended approaches for confirmatory testing of positive Neisseria gonorrhoeae nucleic acid amplification test results. J Clin Microbiol 2008;46:1614–9

LaMontagne

, Fenton

, Pimenta

, Using chlamydia positivity to estimate prevalence: evidence from the Chlamydia Screening Pilot in England. Int J STD AIDS 2005;16:323–7

Scottish Guidelines for Molecular Testing for Neisseria gonorrhoeae . See [http://www.hps.scot.nhs.uk/training/presentations.aspx?id=167] (last accessed 25 February 2009)

10.

Cook

, Hutchison

, Ostergaard

, Systematic review: noninvasive testing for Chlamydia trachomatis and Neisseria gonorrhoeae . Ann Intern Med 2005;142:914–25

11.

Barlow

. Commentary. Sex Transm Infect 2008;84:121

12.

Johnson

REC

, Newhall

, Papp

, Screening tests to detect Chlamydia trachomatis and Neisseria gonorrhoeae infections – 2002. MMWR Recomm Rep 2002;18:1–38

13.

Boyadzhyan

, Yashina

, Yatabe

, Comparison of the Aptima CT and GC assays with the Aptima Combo 2 assay, the Abbott LCx assay, and direct- fluorescent antibody and culture assays for detection of Chlamydia trachomatis and Neisseria gonorrhoeae . J Clin Microbiol 2004;42:3089–93

14.

Golden

, Hughes

, Cles

, Positive predictive value of Gen-Probe APTIMA combo 2 testing for Neisseria gonorrhoeae in a population of women with low prevalence of N. gonorrhoeae infection. Clin Infect Dis 2004;39:1387–90

15.

Moss

, Mallinson

. The contribution of Aptima Combo 2 assay to the diagnosis of gonorrhoea in genitourinary medicine setting. Int. J STD AIDS 2007;18:551–4

16.

O'Mahony

, Fowkes

, Worthen

, Mallinson

. Three years of using Aptima Combo 2(AC2) transcription-mediated amplification for gonorrhoea in a district hospital genitourinary medicine clinic show it to be superior to culture and has a specificity of almost 100%. Int J STD AIDS 2008; 19:67–9

17.

Lister

, Tabrizi

, Fairley

, Validation of Roche Cobas Amplicor assay for detection of Chlamydia trachomatis in rectal and pharyngeal specimens by an omp1 PCR assay. J Clin Microbiol 2004;42:239–41

18.

Alexander

, Martin

, Ison

. Confirming the Chlamydia trachomatis status of referred rectal specimens. Sex Transm Infect 2007;83:327–9

19.

Young

, Anderson

, Moyes

, McMillan

. Non-culture detection of rectal and pharyngeal gonorrhoea by the Gen-Probe PACE 2 assay. Genitourin Med 1997;73:50–62

20.

Alloba

, Allawattegama

, Jones

. Use of nucleic amplification tests for the detection of Neisseria gonorrhoeae: survey of genitourinary medicine clinics in England. Int J STD AIDS 2007;18:652–3

21.

Gonococcal Resistance to Antimicrobials Surveillance Programme (GRASP) Annual Report 2007. See [http://www.hpa.org.uk/web/HPAwebFile/HPAweb_C/1221117895841] (last accessed 25 February 2009)

22.

Ryan

, Kudesia

, McIntyre

, BD ProbeTec ET assay for the diagnosis of gonorrhoea in a high risk population: a protocol for replacing traditional microscopy and culture techniques. Sex Transm Infect 2007; 83:175–9

23.

Lavell

, Jones

, Mallinson

, Finding, confirming, and managing gonorrhoea in a population screened for chlamydia using the Gen-Probe Aptima Combo 2 assay. Sex Transm Infect 2006;82:221–4

24.

Aghaizu

, Atherton

, Mallinson

, Prevalence of Neisseria gonorrhoea infection in young women in South London. See http://sti.bmi.com/cgi/eletters/84/2/117 (last accessed 25 February 2009)

25.

Rao

, Bacon

, Evans

, Prevalence of Neisseria gonorrhoeae infections in young subjects attending community clinics in South London. Sex Transm Infect 2008;84:117–121

26.

John

. Evans-Jones (6 February 2008). The KC60 and the future. E letter available at http://sti.bmj.com/cgi/eletters/83/6/433 (last accessed 25 February 2009)

27.

Lavelle

, Mallinson

, Henning

, Webb

, Hughes

, Abbott

. Impact on Gonorrhoea case reports of concomitant testing in a chlamydia screening population. Sex Transm Infect 2007;83:593–4