Nucleic Acid Amplification of the opa Gene for Detection of Neisseria gonorrhoeae: Experience from a Diagnostic Laboratory

Abstract

We report results of Neisseria gonorrhoeae nucleic acid amplification testing (NAAT) with the Abbott m2000 PCR at a tertiary laboratory 6 months after its introduction. Of 5,475 specimens tested, 45 samples (0.82%) tested positive for N. gonorrhoeae. Eight were not cultured, but seven tested positive with a porA pseudogene NAAT.

Neisseria gonorrhoeae is a sexually transmitted bacterium that causes a spectrum of disease ranging from asymptomatic infection to pelvic inflammatory disease. Gonorrhea is not a notifiable disease in New Zealand, and therefore accurate estimation of the prevalence is difficult. The prevalence has been estimated at 90 cases/100,000 population (2), although this is likely to be an underestimation as it includes only those cases diagnosed at sexual health and family planning clinics. The presumed rate is about 6 times lower than that of chlamydia infection.

Nucleic acid amplification testing (NAAT) was introduced at Canterbury Health Laboratories in addition to culture due to faster turnaround-time (7). Testing is routinely performed with the Abbott m2000 as a duplex PCR for Chlamydia trachomatis and N. gonorrhoeae on all specimens with requests for detection of either microorganism. Since introduction, all specimens with a positive NAAT result have undergone testing with a confirmatory NAAT of the porA pseudogene if they have not been cultured. Routine testing with the Abbott m2000 is performed using Abbott multi-Collect specimen collection kits (MML Diagnostics Packaging, Inc., Troutdale, OR) according to the manufacturer's instructions. The porA pseudogene testing was performed by the method described by Wiley et al. (9). Culture was performed on Thayer-Martin agar (Fort Richard Laboratories, Ltd., New Zealand) in 5% carbon dioxide for 3 days.

Most reports in the literature report results from NAATs of sample collections with a high proportion containing Neisseria gonorrhoeae (5, 7). In contrast, reports from diagnostic use of earlier generation NAATs found low positive predictive values and false positives. Concerns have been raised about the high rates of false-positive results in populations with a low prevalence of N. gonorrhoeae (1, 3, 4, 10).

New Zealand currently conducts NAAT screening for chlamydia in sexually active populations. Addition of N. gonorrhoeae by NAAT means that potentially low-risk population groups would be tested and false-positive results in settings with a low positive predictive value could be detrimental to the screened population.

The objective of this retrospective review was to review cases of N. gonorrhoeae that were diagnosed solely by NAAT and assess the occurrence of likely false-positive results.

All specimens tested by PCR for N. gonorrhoeae between 1 November 2009 and 1 May 2010 were retrospectively reviewed. The study was determined to be exempt from review by the regional ethics committee. Those specimens that had a positive NAAT had the results of their culture or confirmatory NAAT reviewed. The information was analyzed to determine the proportion of specimens with N. gonorrhoeae detected and the proportion of those that were not cultured but had a positive NAAT. Cases in which specimens had N. gonorrhoeae DNA detected but were culture negative were reviewed in further detail by collection of clinical particulars from the request form and treating clinician.

From 1 November 2009 to 1 May 2010, there were 5,475 samples tested. They included 2,684 cervical swabs, 1,544 urine specimens, 856 urethral swabs, 227 anorectal swabs, 100 vaginal swabs, 35 pharyngeal swabs, 27 eye swabs, and 2 pelvic aspirates.

There were 45 specimens with DNA detected by NAAT (0.82% of the specimens tested) and 37 with a positive culture (0.68%). There were 27 positive samples from sexual health clinics, 12 from community general practitioners, 4 from hospital inpatients, and 2 from the hospital emergency department.

There were 45 specimens with a positive N. gonorrhoeae NAAT. Thirty-seven had a concordant positive culture, and 8 specimens did not. When the positive NAAT was known to the medical laboratory scientist, the culture plates were reviewed. Three of the 37 positive cultures were detected during this review. Of the eight samples without a supportive culture following a plate review, three had negative cultures and five did not have a sample submitted for culture. These samples had all been tested with the porA pseudogene NAAT. Seven of eight samples were positive for this gene. All culture-positive samples had N. gonorrhoeae DNA detected by NAAT.

The characteristics of patients with culture-negative, NAAT-positive results are detailed in Table 1.

Table 1.

Clinical and laboratory characteristics of patients with culture-negative, m2000 PCR-positive results

Patient (n = 8)	Gender/age (yr)	Specimen	Result by:		Clinical details
			Abbott m2000 opa gene	porA pseudogene
1	Male/31	Rectal swab	Positive	Positive	Rectal symptoms and previous gonorrhea; delayed inoculation of culture plate; culture negative; presented again 1 month later with culture-proven gonorrhea from rectal swab
2	Female/15	Cervical Swab	Positive	Negative	Pregnant and admitted with preterm labor, C. trachomatis positive, culture negative
3	Female/23	Cervical swab	Positive	Positive	Abdominal pain and purulent discharge, delayed inoculation, culture negative
4	Male/22	Urine	Positive	Positive	Urethritis symptoms; no swab sent for culture
5	Male/25	Urethral	Positive	Positive	Urethritis symptoms; no swab sent for culture
6	Male/23	Urine	Positive	Positive	Dysuria and hematuria; no swab sent for culture
7	Male/18	Urine	Positive	Positive	Penile discharge; no swab sent for culture
8	Female/22	High vaginal swab	Positive	Positive	Fever, lower abdominal pain, and purulent vaginal discharge; no swab sent for culture

The results of this retrospective review show that the rate of N. gonorrhoeae DNA detection that was unsupported by either culture or a second NAAT was extremely low: 1/45 (2.2%). Additionally, this culture-negative group all had either symptoms consistent with infection or a known exposure. All but one specimen with DNA detected that was unsupported by culture had DNA detected by the use of the porA pseudogene. The porA pseudogene has been validated previously as a suitable confirmatory test for positive NAATs (6, 8, 9). The combination of an appropriate clinical history and positive confirmatory NAAT makes a false-positive result very unlikely.

If this group are regarded as true positives, then the positive predicted value of the m2000 system NAAT alone is 97.8% (44/45), using a prevalence of 0.82%. The prevalence rate in the population undergoing testing is 10 times the estimated rate of the general population. This is probably due to specimens coming principally from the sexual health clinic and inpatients. Both of these groups are likely to be at high risk epidemiologically and are more likely to be presenting with symptoms than asymptomatically for screening.

There was one case (case 2) in which the opa gene NAAT was positive on the m2000 but was not confirmed on either culture or porA pseudogene NAAT. The sample was a late positive, in that the difference between the cutoff decision cycle and the specimen's cycle threshold number on the m2000 opa gene NAAT was only 1.0, which was much lower than the mean of 11.5 (2 standard deviations, 5.0 to 18.0). The real-time curve was, however, morphologically similar to those that amplified at earlier cycles. The NAAT discordance may reflect a false positive on opa gene testing or a false negative from porA pseudogene testing, which has been found by some authors to be less sensitive than other confirmatory NAATs (6). A head-to-head comparison of the opa gene on the m2000 and the porA pseudogene method PCR found that the m2000 had a 10-fold greater analytical sensitivity (T. Anderson and M. J. Maze, unpublished data).

Cases 1 and 3 (Table 1) had negative cultures despite two positive NAATs, and both were associated with delayed inoculation of the swab. An additional important finding of this audit was the three cases of gonorrhea that were undiagnosed on initial culture. This shows that NAAT may improve the sensitivity of culture when used in conjunction.

The conclusions reached by this study are limited by being retrospective. The clinician's pretest probability was not assessed in those specimens that did not have DNA detected by NAAT, and in those with DNA detected, clinical information was not comprehensive. Additionally, the findings are applicable only to the opa gene when tested on the m2000 platform.

This retrospective review does not show evidence of false-positive results due to testing of specimens for N. gonorrhoeae by using NAATs. In cases where swabs are logistically difficult to obtain and culture in an optimal time frame, such as in isolated settings, NAAT for the opa gene provides improved sensitivity over culture with high specificity. The addition of a confirmatory PCR for the porA pseudogene further enhances the specificity and consequently the positive predictive value.