Miller, Amanda BBMedSci; Bromhead, Collette PhD; Jones, Mark FRCPA; Tustin, Paul DMLT
From Aotea Pathology Ltd, Wellington, New Zealand.
Conflicts of interest and sources of funding: none declared.
Correspondence: Collette Bromhead, PhD, Aotea Pathology Ltd, 89 Courtenay Place, Wellington, New Zealand. E-mail: email@example.com.
Received for publication February 28, 2012, and accepted April 5, 2012.
Abstract: We compared the performance of cobas 4800 PCR for Neisseria gonorrhoeae with culture and found an association between pipetting errors and positive culture, caused by mucopurulent discharge. The problem may be resolved by pretreating samples with Sputasol, thus improving detection of both Chlamydia trachomatis and Neisseria gonorrhoeae.
Nucleic acid amplification tests (NAATs) are routinely used for detection of Chlamydia trachomatis (CT) in New Zealand; however, traditional culture has remained the method of choice for diagnosing Neisseria gonorrhoeae (NG). This was due to the inadequate specificity of some NAATs and the need for supplementary testing in low prevalence populations.1
In 2009, we installed the cobas 4800 platform (Roche Diagnostics, New Zealand) and began using the CT/NG test initially for CT polymerase chain reaction (PCR) only. This test system is made up of 2 components, the cobas ×480, which performs automated DNA extraction and PCR setup, whereas the cobas Z480 is a real-time PCR amplification platform. The cobas 4800 CT/NG Test includes 2 target sequences for CT, including the cryptic plasmid and major outer membrane protein genes. The new NG assay targets 2 areas of a direct repeat (DR) region called DR-9, which is repeated 3 times on the NG genome and has 2 highly conserved sequence variations.2 This dual target approach improves on the poor specificity of predecessor assays, as has been demonstrated by recent studies,3,4 and this led us to evaluate its performance against our current culture method for NG.
This analysis revealed an association between patients with gonorrhea and failed PCR tests caused by pipetting errors on the cobas ×480 instrument. The mucopurulent discharge caused by infection with NG appears to interfere with automated pipetting on the cobas ×480, despite all samples being collected and transported in a lytic buffer before testing.
The cobas 4800 CT/NG Test pack insert indicates that interference may occur with cervical mucus, as well as >1 × 105 leukocytes in both urines and swabs, and that failed results cannot preclude the presence of infection.2 Despite recommendations, our evaluation showed that only a third of patients with failed test results due to pipetting errors or amplification failure were recalled for testing. Effective laboratory diagnosis of both NG and CT is essential, as asymptomatic infection is frequent and untreated infections can lead to serious sequelae.5,6 We would like to highlight the potential significance of pipetting errors on the cobas 4800 and propose a method to resolve them.
In our laboratory, specimens that produce pipetting errors on the cobas ×480 are vortexed and retested,2 whereupon an average of 83% were resolved. During our study period between January 1, 2011 and September 13, 2011, there were 97/27,377 repeatedly failed cobas 4800 results in our laboratory. Of those 62 failed CT specimens that were accompanied by a swab for culture, 14 (22%) grew NG. This was high compared with the 0.7% of all cultures that grew NG during the same period.
As the cause of the pipetting error was initially unclear, we trialled 3 resolution methods on a subset of specimens that were repeatedly failed (pipetting errors) or invalid (amplification failure). Dilution and heating have previously been shown to reduce inhibition in CT NAAT assays,7–9 whereas Sputasol (1.4% dithiothreitol; Oxoid Ltd., Basingstoke, United Kingdom) is a strong reducing agent shown to decrease the viscosity of sputum samples before PCR testing.10 There were 11 specimens available that yielded repeatedly failed or invalid cobas 4800 results. These were treated by each of the 3 chosen resolution methods. Swabs collected in cobas media have 4.2 mL of buffer available, and urines have a total of 8 to 10 mL of buffer and urine available for testing. Due to the minimum specimen requirement of the cobas 4800, 1 mL of each swab and 2 mL of each urine were tested by each of the 3 methods. For dilution, an aliquot of each specimen was added to an equal volume of sterile cobas PCR media in a sterile 13-mL tube, then vortexed for 1 minute before testing. For heating, an aliquot of each specimen was added to a sterile 13-mL tube, heated to 95°C for 5 minutes in a hot air oven, and then vortexed for 1 minute before testing. For Sputasol treatment, an aliquot of each specimen was added to an equal volume of Sputasol in a sterile 13-mL tube, as previously described,10 and then vortexed for 1 minute before testing. Following treatment, the specimens were retested on the cobas 4800 as per normal procedure.
Samples were deemed resolved if a valid result was obtained from the cobas 4800. A total of 10 out of 11 specimens were resolved, of which 6 were found to be negative, 2 CT-positive, 1 NG-positive, and 1 CT/NG-positive. An invalid urine sample from a patient with hematuria was not resolved. Sputasol treatment produced 10 valid results and visibly reduced the discharge in each sample, whereas heating and dilution resolved 8 and 9 samples, respectively. Both of the latter methods missed a positive CT result that was detected after Sputasol treatment. This positive result had a Cp value of 25.4, indicating that low template was not a factor.
A validation of the Sputasol treatment was performed on 100 valid cobas 4800 specimens with known results. The validation included 46 urines, of which 3 were CT/NG-positive, 18 CT-positive, and 25 CT-negative, and 54 swabs, of which 2 were CT/NG-positive, 27 CT-positive, and 25 CT-negative. There was 100% agreement between the pre- and post-treatment results and no significant difference (P = 1) between the threshold cycle values obtained for the positive CT/NG results.
Following implementation of Sputasol treatment for all failed or invalid specimens, we have treated a further 50 specimens, of which 6 were CT-positive and 1 CT/NG–positive, showing the ongoing importance of resolving these results (Table 1).
We have shown that repeatedly failed specimens from the cobas 4800 CT/NG assay may have a high proportion of NG-positive culture results and may be resolved by Sputasol pretreatment without affecting the performance of the test. Mucus in swabs is noted as a cause of pipetting errors on the cobas 4800, and the manufacturer recommends removing the swab, vortexing, and retesting to resolve.2 Since 2010, our standard practice has been to remove all swabs and vortex before starting testing, which has reduced failed results by 50% without compromising assay performance (data not shown). This may explain why our rate of failed results (0.35%) is low in comparison with a previous report of 0.6% by Rockett et al.3 However, this step does not reduce pipetting errors from urines, and during the study period, we found that all failed results accompanied by a positive NG culture were attributable to male urine specimens.
Our validation data showed that pipetting errors in both swab and urine specimens may be resolved by treatment with Sputasol before retest. In addition to decreasing the viscosity of sputum samples, Sputasol has been shown to increase the recovery of nucleic acids.10 In this study, treatment of specimens before retesting resolved the pipetting errors without affecting the threshold cycle values of the positive results, despite the dilution factor. We expect treatment with Sputasol to improve detection of CT and NG in our laboratory.
Users of the cobas 4800 would benefit from further information in the package insert regarding the causes of pipetting errors in urines and how to resolve failed results. We recommend that laboratories consider using Sputasol to resolve specimens that yield failed or invalid results on the cobas 4800. Wherever pipetting errors persist, referring physicians should be advised that such results require clinical follow-up.
1. Smith DW, Tapsall JW, Lum G. Guidelines for the use and interpretation of nucleic acid detection tests for Neisseria gonorrhoeae
in Australia: A position paper on behalf of the Public Health Laboratory Network. Commun Dis Intell 2005; 29:358–365.
2. Cobas 4800 CT/NG Test: V4.0 [package insert]. Branchburg, NJ: Roche Molecular Systems Inc.; 2011.
3. Rockett R, Goire N, Limnios A, et al.. Evaluation of the cobas 4800 CT/NG test for detecting Chlamydia trachomatis
and Neisseria gonorrhoeae
. Sex Transm Infect 2011; 86:470–473.
4. Tabrizi SN, Unemo M, Limnios AE, et al.. Evaluation of six commercial nucleic acid amplification tests for the detection of Neisseria gonorrhoeae
and other Neisseria
species. J Clin Microbiol 2011; 49:3610–3615.
5. Champoux JJ. Chlamydia. In: Sherris JC, ed. Medical Microbiology: An Introduction to Infectious Diseases, 2nd ed. New York, NY: Elsevier Science Publishing Co, 1990:463–471.
6. Robinson A. Chlamydia trachomatis
—the case for screening. In: Moss TR, ed. International Handbook of Chlamydia, 2nd ed. England, United Kingdom: Euromed Communications Ltd, 2006:13–25.
7. Mahony J, Chong S, Jang D, et al.. Urine specimens from pregnant and nonpregnant women inhibitory to amplification of Chlamydia trachomatis
nucleic acid by PCR, ligase chain reaction, and transcription-mediated amplification: Identification of urinary substances associated with inhibition and removal of inhibitory activity. J Clin Microbiol 1998; 36:3122–3126.
8. Toye B, Woods W, Bobrowska M, et al.. Inhibition of PCR in genital and urine specimens submitted for Chlamydia trachomatis
testing. J Clin Microbiol 1998; 36:2356–2358.
9. Verkooyen RP, Luijendijk A, Huisman WM, et al.. Detection of PCR inhibitiors in cervical specimens by using the AMPLICOR Chlamydia trachomatis
assay. J Clin Microbiol 1996; 34:3072–3074.
10. Baxter CG, Jones AM, Webb K, et al.. Homogenisation of cystic fibrosis sputum by sonication—an essential step for Aspergillus
PCR. J Microbiol methods 2011; 85:75–81.