Abstract
A total of 3,023 clinical specimens were tested for Neisseria gonorrhoeae by using COBAS AMPLICOR (CA) PCR and confirmation of positives by N. gonorrhoeae-specific 16S rRNA PCR. The sensitivity of CA plus 16S rRNA PCR was 98.8%, compared to 68.2% for culture. Confirmation of CA positives increased the positive predictive value from 54.8 to 96.6%.
For the detection of Neisseria gonorrhoeae, culture is still used by many laboratories. However, recent studies showed that nucleic acid amplification-based N. gonorrhoeae tests (NAATs) are much more sensitive (1, 2, 4, 6, 7, 10). The Roche COBAS AMPLICOR (CA) N. gonorrhoeae test is probably the most widely used amplification test for N. gonorrhoeae. It was shown that certain strains of N. subflava, N. cinerea, N. flavescens, N. lactamica, N. sicca, and Lactobacillus species may produce false-positive results with the CA N. gonorrhoeae test (3, 4, 5, 9, and Manual for COBAS AMPLICOR Neisseria gonorrhoeae PCR test, version 1.0, Roche Diagnostics) and some other N. gonorrhoeae NAATs (8). Therefore, it is clear that positive results obtained with the CA N. gonorrhoeae test should be confirmed by another test method (2, 3, 4, 8, 9). In the present study we evaluated the reliability of N. gonorrhoeae testing in a low-prevalence population, using the CA N. gonorrhoeae test as a screening assay and an N. gonorrhoeae-specific 16S rRNA PCR for confirmation.
N. gonorrhoeae PCR and culture were performed on 3,023 specimens from nonselected patients (2,415 female and 608 male) obtained by visiting general practitioners (48%), general hospitals (32%), or venereal disease (VD) clinics (20%). The vast majority of the patients were symptomatic, and 75% of them were between 15 and 35 years of age. Over 95% of the specimens were urogenital swabs, and the remaining 5% were rectal, throat, or eye swabs. Swabs for PCR and culture were sent to the laboratory in 2-sucrose-phosphate medium and in charcoal transport medium, respectively. The CA N. gonorrhoeae PCR (Roche) was tested according to the manufacturer's instructions (Roche Diagnostics manual, version 1). A specimen was considered negative for N. gonorrhoeae if the optical density (OD) was <0.200 and the OD of the inhibition control (IC) was ≥0.200. The specimen was considered positive for N. gonorrhoeae if the OD was ≥0.200, regardless of the IC result. All specimens with an initial positive result were retested in the CA N. gonorrhoeae PCR, and the result of this repeat CA test was conclusive for the final interpretation of the CA N. gonorrhoeae test. For the 16S rRNA N. gonorrhoeae PCR test, new DNA extracts from the original clinical specimens were prepared. The 16S rRNA N. gonorrhoeae PCR test (Roche) was also performed according to the manufacturer's instructions (manual for Neisseria gonorrhoeae 16S rRNA PCR test, version 2, Roche Diagnostics) and included hybridization with an N. gonorrhoeae-specific probe added to microwell plates. Specimens repeatedly positive by CA N. gonorrhoeae PCR and positive by 16S rRNA N. gonorrhoeae PCR were considered to be N. gonorrhoeae PCR positive. All other specimens were N. gonorrhoeae PCR negative. For culture, the specimens were inoculated on chocolate agar and chocolate agar with vancomycin-trimethoprim-colistin-sulfate-amphotericin. The media were incubated at 37°C with 5 to 10% CO2 for 2 days and then were examined for N. gonorrhoeae-like colonies. Suspected colonies were further examined by an oxidase test and gram staining and were also subcultured. The results of all PCR tests were compared with the results of N. gonorrhoeae culture. Discordant specimens, which were N. gonorrhoeae PCR positive and culture negative, were further investigated using the ligase chain reaction (LCR) N. gonorrhoeae test (Abbott Laboratories). True N. gonorrhoeae-positive specimens were all specimens which were positive by N. gonorrhoeae culture as well as those additional specimens which were repeatedly positive by the CA N. gonorrhoeae PCR, and by 16S rRNA N. gonorrhoeae PCR, and by the LCR N. gonorrhoeae assay. All other specimens were defined as true N. gonorrhoeae negative.
A total of 190 out of 3,023 (6.3%) specimens were initially positive by CA. After retesting, only 155 out of 190 (81.6%) of these remained positive. The reason for this discrepancy is unknown. A similar lack of reproducibility was described earlier for CA Chlamydia trachomatis PCR (11). Of the 155 repeatedly CA N. gonorrhoeae PCR-positive specimens, 87 (56.1%) were also positive by the 16S rRNA N. gonorrhoeae PCR. Only 57 (65.5%) of these 87 N. gonorrhoeae PCR positives were culture positive. In addition, one of the N. gonorrhoeae PCR-negative samples gave a positive culture. Further analyses of the 30 samples that were N. gonorrhoeae PCR positive and N. gonorrhoeae culture negative by LCR resulted in 27 additional true-positive specimens.
The overall prevalence of N. gonorrhoeae in this study was 2.8%, but with urogenital specimens there were large differences between women (1.2%) and men (8.9%) (P < 0.001, Pearson's chi-squared test). Table 1 shows the overall sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of each method. Without confirmation by 16S rRNA N. gonorrhoeae PCR, the PPV of CA N. gonorrhoeae PCR on all samples was only 54.8%. For urogenital specimens from women and men, the PPV of the CA N. gonorrhoeae PCR was 43.3 and 96.0%, respectively. Recently, Diemert et al. (3) showed that additional confirmatory PCR was necessary by comparing N. gonorrhoeae culture and CA N. gonorrhoeae PCR in 737 samples in a very-low-prevalence setting of 0.5%. However, they did not further investigate the validity of their 16S rRNA N. gonorrhoeae PCR confirmation assay for the discordant specimens that were N. gonorrhoeae PCR positive but culture negative. Martin et al. (6) and Van Der Pol et al. (9) described another algorithm to increase the PPV of the CA N. gonorrhoeae test: retesting in duplicate, depending on the initial OD value. When we applied this algorithm to our data, the PPV of the CA N. gonorrhoeae PCR was 65.9%, which is much lower than the PPV of at least >90% described by Van Der Pol et al. (9).
TABLE 1.
Test characteristics
| Specimen typec and/or criterion | Value for test method
|
||
|---|---|---|---|
| N. gonorrhoeae culture | CA N. gonorrhoeae PCRa | Final N. gonorrhoeae PCR resultb | |
| True positive | |||
| No. of positive results | 58 | 85 | 84 |
| No. of negative results | 27 | 0 | 1 |
| True negative | |||
| No. of positive results | 0 | 70 | 3 |
| No. of negative results | 2,938 | 2,868 | 2,935 |
| Sensitivity (%) | 68.2 | 100 | 98.8 |
| Specificity (%) | 100 | 97.6 | 99.9 |
| Positive predictive value (%) | 100 | 54.8 | 96.6 |
| Negative predictive value (%) | 99.1 | 100 | 100 |
Only specimens which were repeatedly reactive in the CA N. gonorrhoeae PCR test were considered to be CA N. gonorrhoeae PCR positive.
Specimens which were repeatedly positive by CA N. gonorrhoeae PCR and positive by 16SrRNA N. gonorrhoeae PCR were considered to be N. gonorrhoeae PCR positive. All other specimens were N. gonorrhoeae PCR negative.
Specimens were considered to be true N. gonorrhoeae positive or true N. gonorrhoeae negative as described in the text.
Of the 70 CA N. gonorrhoeae PCR false-positive specimens, 30 (42.9%) were not obtained from the urogenital tract but were mainly from throat specimens (n = 21). The inclusion of these nonurogenital specimens influenced the PPV of the CA N. gonorrhoeae PCR, being 65.8% for urogenital specimens and 21.1% for nonurogenital specimens. The most likely explanation for the high percentage of false positives with throat swabs is cross-reaction with N. cinerea and N. subflava (3, 4, 5, 9, and Roche Diagnostics manual, version 1). Most of the remaining false-positive reactions may be explained by cross-reactions with Lactobacillus species, which belong to the normal flora of the female genital tract (4).
The present study shows that CA N. gonorrhoeae PCR plus 16S rRNA N. gonorrhoeae PCR confirmation is much more sensitive (98.8%) than culture (68.2%) (P < 0.001, Fisher's exact test). Moreover, the enormous surplus of positive results was very specific, as was shown by the additional confirmation by LCR. The low sensitivity of culture compared to that of PCR could not be explained by suboptimal transport conditions, because culture results obtained with specimens collected very close to the laboratory did not significantly differ from the results obtained with specimens collected at much larger distances (data not shown).
Because the sensitivity of culture might be dependent on the presence of clinical symptoms, we investigated the clinical data of all 18 N. gonorrhoeae-positive patients from one of our hospitals. It appeared that 7 of 8 (87.5%) patients with clinical symptoms and only 2 of 10 (20%) patients without symptoms were N. gonorrhoeae culture positive (P = 0.004, Pearson's chi-squared test).
From 18 N. gonorrhoeae PCR-positive patients we received a follow-up specimen 13 to 35 days (average, 24 days) after antibiotic treatment. None of these follow-up specimens were positive for N. gonorrhoeae by PCR. Therefore, N. gonorrhoeae PCR can also be used for follow-up treatment. Only when susceptibility testing is required does N. gonorrhoeae culture remain necessary.
In conclusion, due to its high sensitivity, N. gonorrhoeae PCR (or perhaps another NAAT) should replace N. gonorrhoeae culture, especially when asymptomatic patients are tested. CA N. gonorrhoeae PCR is a very useful screening assay due to its semiautomatic platform, making it much easier to perform bulk testing, and to its ability to test for Chlamydia trachomatis simultaneously. However, CA N. gonorrhoeae PCR-positive results should always be confirmed with an alternative PCR test, such as the 16S rRNA PCR, especially when low-prevalence populations or nonurogenital specimens are tested.
Acknowledgments
We thank S. Bruisten (GG and GD, Amsterdam, The Netherlands) for performing the LCR assay on discrepant specimens. We also thank J. Moedt and I. K. Roozeboom-Roelfsema for excellent technical assistance.
REFERENCES
- 1.Carroll, K. C., W. E. Aldeen, M. Morrison, R. Anderson, D. Lee, and S. Mottice. 1998. Evaluation of the Abbott Lcx ligase chain reaction assay for the detection of Chlamydia trachomatis and Neisseria gonorrhoeae in urine and genital swab specimens from a sexually transmitted disease clinic population. J. Clin. Microbiol. 36:1630-1633. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Crotchfeld, K. A., L. E. Welsh, D. DeBonville, M. Rosenstraus, and T. C. Quinn. 1997. Detection of Neisseria gonorrhoeae and Chlamydia trachomatis in genitourinary specimens from men and women by a coamplification PCR assay. J. Clin. Microbiol. 35:1536-1540. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Diemert, D. J., M. D. Libman, and P. Lebel. 2002. Confirmation by 16S rRNA PCR of the COBAS AMPLICOR CT/NG test for diagnosis of Neisseria gonorrhoeae infection in a low-prevalence population. J. Clin. Microbiol. 40:4056-4059. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Doornum, G. J. J., L. M. Schouls, A. Pijl, I. Cairo, M. Buimer, and S. Bruisten. 2001. Comparison between the LCx probe system and the COBAS AMPLICOR system for detection of Chlamydia trachomatis and Neisseria gonorrhoeae infections in patients attending a clinic for treatment of sexually transmitted diseases in Amsterdam, The Netherlands. J. Clin. Microbiol. 39:829-835. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Farrell, D. J. 1999. Evaluation of AMPLICOR Neisseria gonorrhoeae PCR using cppB nested PCR and 16S rRNA PCR. J. Clin. Microbiol. 37:386-390. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Martin, D. H., C. Cammarata, B. Van Der Pol, R. B. Jones, T. C. Quinn, C. A. Gaydos, K. Crotchfelt, J. Schachter, J. Moncada, D. Jungkind, B. Turner, and C. Peyton. 2000. Multicenter evaluation of AMPLICOR and automated COBAS AMPLICOR CT/NG tests for Neisseria gonorrhoeae. J. Clin. Microbiol. 38:3544-3549. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Moncada, J., J. Schachter, E. W. Hook III, D. Ferrero, C. Gaydos, T. C. Quinn, D. Willis, A. Weissfeld, and D. H. Martin. 2004. The effect of urine testing in evaluations of the sensitivity of the Gen-Probe APTIMA Combo 2 Assay on endocervical swabs for Chlamydia trachomatis and Neisseria gonorrhoeae. Sex. Trans. Dis. 31:273-277. [DOI] [PubMed] [Google Scholar]
- 8.Palmer, H. M., H. Mallinson, R. L. Wood, and A. J. Herring. 2003. Evaluation of the specificities of five DNA amplification methods for the detection of Neisseria gonorrhoeae. J. Clin. Microbiol. 41:835-837. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Van Der Pol, B., D. H. Martin, J. Schachter, T. C. Quinn, C. A. Gaydos, R. B. Jones, K. Crotchfelt, J. Moncada, D. Jungkind, B. Turner, C. Peyton, J. F. Kelly, J. B. Weiss, and M. Rosenstraus. 2001. Enhancing the specificity of the COBAS AMPLICOR CT/NG tests for Neisseria gonorrhoeae by retesting specimens with equivocal results. J. Clin. Microbiol. 39:3092-3098. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Van Der Pol, B., D. V. Ferrero, L. Buck-Barrington, E. Hook III, C. Lenderman, T. C. Quinn, C. A. Gaydos, J. Lovchik, J. Schachter, J. Moncada, G. Hall, M. J. Tuohy, and R. B. Jones. 2001. Multicenter evaluation of the BDProbeTec ET system for detection of Chlamydia trachomatis and Neisseria gonorrhoeae in urine specimens, female endocervical swabs, and male urethral swabs. J. Clin. Microbiol. 39:1008-1016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Vincelette, J., J. Schirm, M. Bogard, A.-M. Bourgault, D. S. Luijt, A. Bianchi, P. C. van Voorst Vader, A. Butcher, and M. Rosenstraus. 1999. Multicenter evaluation of the fully automated COBAS AMPLICOR PCR test for detection of Chlamydia trachomatis in urogenital specimens. J. Clin. Microbiol. 37:74-80. [DOI] [PMC free article] [PubMed] [Google Scholar]