Literature review > Issue_4 > Review on Chong et al. 

The introduction of nucleic acid amplification tests (NAAT) has revolutionized testing in reference laboratories for chlamydial infection. In addition to their remarkable sensitivity, they also provide very high specificity. They can effectively be used in asymptomatic populations for case finding and with relatively noninvasive specimens such as exudates or urines [1]. However, one downside of NAAT is the presence of inhibitory substances in some specimens. Several reports have described inhibitors that can lead to false-negative results [2-7]. This report by Chong et al., is one of several from this research group investigating and documenting the presence of inhibitors. These include NAAT that are or have been routinely in use, such as the Roche Amplicor PCR, Abbott Chlamydia LCX, and GenProbe Chlamydia TMA.

Like previous reports from this group, this study also detected inhibitors by "spiking" normal urines with cultured chlamydial organisms and that inhibitors were inactivated by overnight storage or by dilution. Unlike the previous reports, however, this report did not characterize the nature of the inhibitors, found no difference in the frequency of inhibitors in urines from pregnant vs. non-pregnant women, and found 13% false negative results with the LCx, which is significantly higher compared with 2.6% in a previous publication [4].

In this study an input concentration of one (1) chlamydial elementary body (EB) per sample was claimed, but the preparations must have surely underestimated the numbers of EB used and/or may also have included co-purified free target nucleic acids. To this regard, the authors detected only seven false-negatives in 244 urines in the APTIMA, a sensitivity of 97.1%. This result is statistically highly unlikely, since the Poisson distribution [8] predicts that where there is a random mean input of one unit (one EB) per volume (100 microliters) of sample, only 63% of the assays would receive one EB or more. For this study of 244 samples, assuming that a minimum of 1 EB is required to produce a positive test, approximately 154 positive and 90 negative results would have been expected. Also note that in this study, a stock solution of EB was diluted 10-fold from 10^-0 to 10^-8. Using direct immunofluorescence, the mean number of EB detected was 12 EB per 100 microliters in the 10^-5 dilution and one EB in the 10^-7 dilution, which is only one-twelfth the number of EB in a solution that was diluted 100-fold!

It is also unclear why, in experiment 1 of this study, 13% of the specimens tested as "false negative" by the LCX, whereas in experiment 2, 48.6%, or nearly four times that of first experiment, initially tested "false negative" and then remained negative. Was there a difference in the nature of the urines or concentration of the chlamydial EB? The authors did not explain this apparent discrepancy. The authors also did not use formal statistical testing to draw conclusions in this report. These analyses should have been performed, given that it is fairly simple to compare two proportions formally with either a Pearson-chi squared test for unpaired data (e.g., pregnant vs. non-pregnant women) or McNemar's test for paired data (e.g., specimens from the same subject before and after processing).
It may also have been more productive to design a study in which sensitivity (or false negativity) is expressed as a function of the input numbers of target nucleic acid molecules in the sample. The influence of NAAT inhibitors in urine can then be accurately measured. For example, in a Mycoplasma genitalium study, Hughes and Totten [9] used serially diluted aliquots of a known concentration of the target DNA to determine test sensitivity.

In summary, this report compares the performance of the Abbott LCX with the GenProbe APTIMA Combo 2 test, concluding that the APTIMA is at least 100-fold more sensitive than the LCX, less influenced by inhibitors, and less susceptible to loss of target DNA in specimen processing. This conclusion may not be that surprising, considering that the APTIMA uses signal amplification of 16S ribosomal RNA, which can be found in much higher copy number per organism when compared with target amplification of genomic DNA for the LCx, giving the APTIMA an advantage when the input numbers of EB are low.

Unfortunately, the performance of the APTIMA compared to the LCx, and the potential applicability of these findings to better inform potential and current users of NAAT is currently somewhat of a moot point. Abbott Laboratories discontinued production of their pioneering LCX test in 2003.

References:

1. Johnson RE, Green TA, Schachter J, Jones RB, Hook EW 3rd, Black CM, Martin DH, St Louis ME, Stamm WE. Evaluation of nucleic acid amplification tests as reference tests for Chlamydia trachomatis infections in asymptomatic men. J Clin Microbiol. 2000 Dec;38(12):4382-6.

2. Chernesky M, Chong S, Jang D, Luinstra K, Faught M, Mahony J Inhibition of amplification of Chlamydia trachomatis plasmid DNA by the ligase chain reaction associated with female urines. Clin Microbiol Infect. 1998; 4(7): 397-404.

3. Chernesky MA, Jang D, Sellors J, Luinstra K, Chong S, Castriciano S, Mahony JB. Urinary inhibitors of polymerase chain reaction and ligase chain reaction and testing of multiple specimens may contribute to lower assay sensitivities for diagnosing Chlamydia trachomatis infected women. Mol Cell Probes. 1997; 11(4): 243-9

4. Mahony J, Chong S, Jang D, Luinstra K, Faught M, Dalby D, Sellors J, Chernesky M. 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(11): 3122-6.

5. Castriciano S, Luinstra K, Jang D, Patel J, Mahony J, Kapala J, Chernesky M. Accuracy of results obtained by performing a second ligase chain reaction assay and PCR analysis on urine samples with positive or near-cutoff results in the LCx test for Chlamydia trachomatis. J Clin Microbiol. 2002 Jul;40(7):2632-4.

6. Berg ES, Anestad G, Moi H, Storvold G, Skaug K. False-negative results of a ligase chain reaction assay to detect Chlamydia trachomatis due to inhibitors in urine. Eur J Clin Microbiol Infect Dis. 1997 Oct;16(10):727-31.

7. Pasternack R, Vuorinen P, Miettinen A. Evaluation of the Gen-Probe Chlamydia trachomatis transcription-mediated amplification assay with urine specimens from women. J Clin Microbiol. 1997 Mar;35(3):676-8.

8. Taswell C. Limiting dilution assays for the determination of immunocompetent cell frequencies. Journal of Immunology 1981; 126: 1614-1619

9. Hughes JP, Totten P. Estimating the Accuracy of Polymerase Chain Reaction-Based Tests Using Endpoint Dilution. Biometrics 2003; 59(9), 505-511

about SDI | newsletters | grants | publications | literature reviews

WHO Home - WHO Search - TDR Home - SDI Home - SDI Contact us
(c) WHO/OMS 2001