This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Cloud, J. L.
Right arrow Articles by Woods, G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Cloud, J. L.
Right arrow Articles by Woods, G.

 Previous Article  |  Next Article 

Journal of Clinical Microbiology, November 2004, p. 5341-5344, Vol. 42, No. 11
0095-1137/04/$08.00+0     DOI: 10.1128/JCM.42.11.5341-5344.2004
Copyright © 2004, American Society for Microbiology. All Rights Reserved.

Evaluation of a Modified Gen-Probe Amplified Direct Test for Detection of Mycobacterium tuberculosis Complex Organisms in Cerebrospinal Fluid

Joann L. Cloud,1* Cheryl Shutt,1 Wade Aldous,1,2 and Gail Woods1,2

ARUP Institute for Clinical and Experimental Pathology,1 Department of Pathology, University of Utah, Salt Lake City, Utah2

Received 14 April 2004/ Returned for modification 10 June 2004/ Accepted 27 July 2004


arrow
ABSTRACT
 
Laboratory evidence for tuberculous meningitis is difficult to acquire due to the low numbers of organisms present in cerebrospinal fluid (CSF) and the presence of nucleic acid amplification inhibitors. The Amplified Mycobacterium tuberculosis Direct Test (MTD) is sensitive and specific for the direct detection of M. tuberculosis complex in respiratory samples but has not been approved for CSF. We evaluated a modified version of the current MTD, optimized for use with CSF samples. Samples were prepared by spiking CSF with various numbers of M. tuberculosis complex organisms. The modified MTD performance was compared with results obtained using a purified RNA sample extracted using the Qiagen RNeasy Protect Bacteria Mini Kit. By use of CSF artificially spiked with M. tuberculosis complex, the sensitivity of the modified MTD was 100% (six of six) for CSF samples containing approximately 600 CFU/ml, 78% (seven of nine) for approximately 60 CFU/ml, 50% (three of six) for 6 CFU/ml, and 17% (one of six) for samples with <1 CFU/ml. The specificity of the modified MTD method was 100% (22 of 22). The sensitivity of the Qiagen MTD method was 100% for CSF samples containing approximately 600 CFU/ml (six of six) and approximately 60 CFU/ml (nine of nine), 50% for samples with approximately 6 CFU/ml (three of six), and 50% for samples with <1 CFU/ml (three of six). The specificity of the Qiagen MTD method was 86% (19 of 22). With the Qiagen MTD method, however, initial results were equivocal for 14 of the 27 (52%) positive samples, requiring repeat analysis, whereas with the modified MTD, only 1 of 27 (4%) was equivocal. The modified MTD for CSF samples was less time-consuming and less expensive and resulted in considerably fewer equivocal results than the Qiagen MTD method did.


arrow
TEXT
 
Meningitis is one of the most devastating manifestations of tuberculosis. Unfortunately, the diagnosis of tuberculous meningitis (TBM) is difficult due to the low number of bacteria in cerebrospinal fluid (CSF), leading to poor sensitivity of acid-fast staining and culture (4). Given its ability to provide rapid diagnosis and its potential for increased sensitivity, many investigators have examined the use of nucleic acid amplification (NAA), both commercial and "home brew," for diagnosis of TBM (2, 3, 5-7, 9). Currently, two commercial NAA tests for direct detection of Mycobacterium tuberculosis complex (MTBC) are marketed in the United States: Amplified Mycobacterium tuberculosis Direct Test (MTD; Gen-Probe, Inc., San Diego, Calif.) and AMPLICOR Mycobacterium tuberculosis Test (Roche Diagnostic Systems, Inc., Indianapolis, Ind.). MTD is approved by the Food and Drug Administration for testing respiratory specimens, regardless of the result of the smear for acid-fast bacilli (AFB); AMPLICOR is approved for testing AFB smear-positive specimens only. Neither test is approved for testing CSF.

The sensitivity of detection of MTBC in CSF is increased, but not ideal, with NAA tests (5, 7). To improve the sensitivity of NAA tests, the problems of amplification inhibitors present in CSF and low bacterial load need to be addressed. When using the initial version of MTD to test CSF (artificially spiked with MTBC), Pfyffer et al. showed that the sensitivity of the assay was improved considerably by increasing the sample volume and pretreating it with sodium dodecyl sulfate (SDS), a detergent that denatures protein and enzymes and, therefore, theoretically should remove substances that might inhibit amplification (6). This increased sensitivity associated with SDS pretreatment was recently confirmed by Thwaites et al. when testing spiked CSF with the current (enhanced) version of the MTD (9). A drawback to using SDS with the MTD, however, is the potential to denature the enzyme essential for amplification and consequently adversely affect the test performance. Therefore, we investigated alternative methods of eliminating inhibitory substances: (i) simple dilution of the sample and (ii) purification and concentration of the RNA with the RNeasy Mini Kit (Qiagen, Valencia, Calif.).

A suspension of M. tuberculosis (ATCC 27294) was made in 7H9 broth (Midwest Medical, Salt Lake City, Utah) to equal the density of a 0.5 McFarland standard. Serial 10-fold dilutions to 10–5 were prepared in saline, and colony counts were performed by culturing 100 µl of the 10–4 and 10–5 dilutions on 7H11 medium (Midwest Medical) and incubating them at 37°C for 5 weeks. Aliquots of 200 µl were frozen at –80°C for later use. Dilutions from the frozen stocks were made to artificially spike CSF samples with M. tuberculosis. To simulate samples negative for M. tuberculosis, aliquots of CSF were spiked with Haemophilus influenzae, Staphylococcus aureus, Streptococcus pneumoniae, Escherichia coli, Streptococcus agalactiae, and Cryptococcus neoformans. These organisms were selected because they are among the more common causes of meningitis and are those likely to be encountered clinically. We did not include nontuberculous mycobacteria because they rarely cause meningitis. Additionally, the manufacturer has performed extensive cross-reactivity studies and shown that, with the exception of Mycobacterium celatum, cross-reactivity with nontuberculous mycobacteria does not occur.

CSF samples from patients with diagnoses other than TBM, after storage for 1 month at –20°C at ARUP Laboratories, were pooled to a volume of 1,500 µl and used for spiking purposes. A total of 27 sample pools were spiked with M. tuberculosis by adding 15 µl of each dilution to 1,500 µl of CSF. The 10–2 dilution was added to six CSF pools, the 10–3 dilution was added to nine CSF pools, the 10–4 dilution was added to six CSF pools, and the 10–5 dilution was added to six CSF pools. Twenty-two negative samples were prepared by adding 15 µl of the 10–2 dilution from a 0.5 McFarland standard of each of the non-MTBC strains listed above to 1,500 µl of CSF. The actual number of CFU per milliliter of sample was determined based on results of the colony counts.

All laboratory procedures were performed in a biosafety level 3 laboratory in a biosafety cabinet. The procedure for the modified MTD follows the manufacturer's instructions for the current version of the commercially available kit with the following procedural alterations. A CSF volume of 300 µl (rather than 450 µl for respiratory specimens) was added to 150 µl of PCR-grade water and 50 µl of specimen dilution buffer; the mixture was then placed into the lysing tube. Amplification time was 1 h (30 to 60 min is recommended for respiratory specimens). The remainder of the procedure was in accordance with the manufacturer's instructions. Quality control samples were prepared and tested per the manufacturer's recommendations. Results were interpreted as recommended by the manufacturer for respiratory specimens: <30,000 relative light units (RLU), negative; ≥500,000 RLU, positive; 30,000 to 499,999 RLU, equivocal. Samples yielding equivocal results were retested, using the reserved lysate. A repeat result of ≥30,000 RLU was considered positive for MTBC; a result of <30,000 RLU was negative.

For the Qiagen MTD (i.e., Qiagen extraction followed by the current unaltered MTD), the enzymatic lysis protocol for stabilization and isolation of total RNA from bacterial cultures (RNAprotect Bacteria Reagent handbook, Qiagen, July 2001) was followed. Briefly, a volume of 1,350 µl of RNAprotect Bacteria Reagent (Qiagen) was added to 675 µl of sample to obtain the appropriate ratio of reagent to sample. This pertains to the maximum volume (2 ml) allowable for use with a microcentrifuge. The mixture was vortexed briefly, incubated for 5 min at room temperature, and centrifuged for 10 min at 15,000 x g, after which the supernatant was decanted. The pellet was treated with 15 mg of lysozyme/ml for 10 min at room temperature and then boiled for 15 min to heat-kill the organisms. Thereafter, the RNA extraction procedure described in the RNAprotect Bacteria Reagent handbook was followed, using the RNeasy miniprotocol for isolation of total RNA from bacteria.

After extraction, RNA samples were prepared for the MTD according to the method of Mester et al. (J. Mester, A. Somoskovi, A. Waring, L. M. Parsons, and M. Salfinger, Abstr. 102nd Gen. Meet. Am. Soc. Microbiol., abstr. C-51, 2002). Briefly, 25 µl of the Qiagen product was added to 92.5 µl of sterile water and 12.5 µl of Gen-Probe lysis buffer. From this purified sample mixture, 25 µl was added to the amplification reaction mixture. The remaining steps of the MTD procedure were followed per the manufacturer's instructions for the current version.

The study was conducted in two parts. In phase 1, current MTD and modified MTD were compared to assess the necessity of altering the test. An aliquot of the frozen M. tuberculosis 0.5 McFarland suspensions (see above) was used to make five 10-fold dilutions (10–4 to 10–7) in CSF. All dilutions were tested by the unaltered current MTD and the modified MTD. The limit of detection was determined for each method. In phase 2, the modified MTD and the Qiagen MTD were compared by testing the 49 spiked samples (27 with M. tuberculosis and 22 with other organisms) by both methods. All samples were intersorted and blinded to the person performing the assays.

Based on our dilution studies, the 10–4 and 10–5 dilutions of the original 0.5 McFarland suspension revealed an average of 600 and 60 CFU/ml, respectively (data not shown). From these data, we calculated that the 10–6 and 10–7 dilutions contained 6 and <1 CFU/ml, respectively.

In the first phase of the study the limits of detection of the unaltered current MTD and the modified MTD were 600 and 60 CFU/ml, respectively. Results suggest that the lower limit of detection of the modified MTD likely is due to decreased inhibitors in the sample and confirm the increased sensitivity of the modified version.

Results of phase 2 of the evaluation, in which 49 artificially spiked CSF samples were analyzed by both the modified MTD and the Qiagen MTD, are summarized in Table 1. By use of the modified MTD, 16 (59%) of the 27 samples spiked with M. tuberculosis were positive. Six of six (100%) samples with 600 CFU/ml, seven of nine (78%) with 60 CFU/ml, and three of six (50%) with 6 CFU/ml were positive by the modified MTD. Of the six samples spiked with <1 CFU/ml, only one was positive (equivocal on initial testing). All samples (22 of 22) spiked with organisms other than M. tuberculosis were negative by the modified MTD (specificity, 100%).


View this table:
[in this window]
[in a new window]
  		TABLE 1. RLU for all samples tested by the modified MTD and Qiagen MTD methodsa

With the Qiagen MTD method, 100% (six of six) of samples with 600 CFU/ml, 100% (nine of nine) with 60 CFU/ml, 50% (three of six) with 6 CFU/ml, and 50% (three of six) with <1 CFU/ml were positive. Of all M. tuberculosis-spiked samples, 21 of 27 (78%) were positive by Qiagen MTD; 13 of the 21 were within the equivocal range upon initial testing. All Qiagen MTD-equivocal samples were positive (≥30,000 RLU) after repeat analysis. Of the negative samples, 19 of 22 resulted in <30,000 RLU and three samples showed >500,000 RLU (false positive) by Qiagen MTD (specificity, 86%).

The laboratory diagnosis of TBM involves CSF white blood cell count and protein and glucose concentrations, detection of AFB in the CSF by microscopy of direct smears, and mycobacterial culture of CSF in a patient with an appropriate clinical presentation. Culture, often performed routinely, has a sensitivity of about 50% (11), and results often are not available for weeks. Microscopy is rapid and inexpensive but has a sensitivity of only 10 to 50% and a limit of detection of 104 to 105 organisms/ml (8-11). The sensitivity of microscopy varies extensively depending on the volume of CSF examined, the duration of microscopy, and the skill of the microscopist. CSF white blood cell count, protein, and glucose values generally are available within hours, but results are not specific for TBM. Given the diagnostic limitations of these tests, many investigators have examined the use of NAA, amplifying target nucleic acid regions that identify MTBC. In most of these studies, however, the sensitivity (and often specificity) of NAA has been lower when testing CSF than when testing respiratory specimens. Reasons for the decreased sensitivity are the lower number of organisms and the presence of substances that inhibit amplification.

To eliminate inhibitory substances when using MTD to test CSF, pretreatment of the sample with SDS has been suggested (6). Use of SDS, however, not only would denature proteins and enzymes that inhibit amplification but also could denature the enzyme essential for amplification if not entirely removed from the pretreated sample by thorough washing. If any SDS remained, MTD performance could be compromised. This could explain the low sensitivity of MTD in a recent study of TBM reported by Thwaites et al. (11). Given this potential drawback of SDS, we investigated other means to eliminate inhibitory substances. We chose sample dilution based on its technical simplicity and personal experience with other amplification platforms. We also evaluated the RNAprotect RNeasy kit (Qiagen), although it adds approximately 2 h to the test turnaround time, because it should eliminate inhibitors and concentrate the sample to potentially enhance the sensitivity.

The major limitation of our study was the use of artificially spiked CSF rather than clinical specimens. However, it was impractical, if not impossible, for us to use clinical samples for two reasons. First, the number of CSF specimens that are culture positive for MTBC at ARUP is ≤5/yr (<1% of all CSF specimens received). Therefore, if clinical specimens were used, the study would have required years to complete. Second, the vast majority of the CSF samples received at ARUP are from outside clients and consist of <1 ml. This volume is less than what is recommended for culture and is definitely insufficient for the additional testing required for our study.

Our data showed that Qiagen MTD was more sensitive than the modified MTD; however, it also was less specific. With the Qiagen MTD, there were three false-positive results, all >500,000 RLU and two of the three showing >1 million RLU. This likely is due to the greater number of manipulations associated with the extraction, leading to an increased likelihood of cross-contamination. Additionally, the Qiagen MTD gave more equivocal results on initial testing than the modified MTD did (52 versus 4%, respectively), and it was considerably more labor-intensive, requiring approximately 2 h to perform the extraction. Both the high number of equivocal results and the added technical time increase the cost of the test considerably.

One aspect of the test that could not be adequately addressed with spiked samples was cutoff for a positive result. We chose the conservative approach, using the manufacturer's recommendations for respiratory specimens. With the one clinical sample that we tested, those recommendations were appropriate. Other investigators have used or suggested different cutoff values. Baker et al. (1) used the MTD without modification and performed testing in duplicate; if at least one of the assays yielded a result of ≥30,000 RLU, it was considered positive. Lang et al. (5), using the initial version of the MTD, found that a cutoff of ≥11,000 RLU considerably increased sensitivity without decreasing specificity. Our data showed that, with the modified MTD, a cutoff of ≥11,000 RLU would not be appropriate because two specimens would have given false-positive results. One could consider changing the lower limit of the equivocal zone to 20,000 RLU; however, many more clinical samples must be tested before such a recommendation could be made.

In summary, if laboratories that currently use MTD for testing respiratory specimens are asked to test CSF, using the modification of the test described here is reasonable. However, because the MTD is not Food and Drug Administration-approved for testing CSF, a disclaimer should be added to the report indicating in-house validation of the protocol. The modified MTD procedure provides a good limit of detection and requires no added costs or labor over the existing MTD. The Qiagen extraction is another option and may increase sensitivity, but extreme care must be taken to minimize the possibility of contamination. For interpretation of results, we recommend the conservative approach, using the manufacturer's guidelines for respiratory specimens. However, more experience testing clinical specimens may show that a lower cutoff is acceptable.


arrow
ACKNOWLEDGMENTS
 
This work was supported by the ARUP Institute for Clinical and Experimental Pathology.

The MTD kits were kindly provided by Gen-Probe, Inc. We are grateful to the ARUP Microbiology Laboratory for the use of their biosafety level 3 facilities.


arrow
FOOTNOTES
 
* Corresponding author. Mailing address: ARUP Institute for Clinical and Experimental Pathology, 500 Chipeta Way, Salt Lake City, UT 84108. Phone: (801) 583-2787, ext. 2439. Fax: (801) 584-5207. E-mail: cloudjl{at}aruplab.com. Back


arrow
REFERENCES
 
    1
  1. Baker, C. A., C. P. Cartwright, D. N. Williams, S. M. Nelson, and P. K. Peterson. 2002. Early detection of central nervous system tuberculosis with the Gen-Probe nucleic acid amplification assay: utility in an inner city hospital. Clin. Infect. Dis. 35:339-342.[Medline]
    2. 2
  2. Caws, M., S. M. Wilson, C. Clough, and F. Drobniewski. 2000. Role of IS6110-targeted PCR, culture, biochemical, clinical, and immunological criteria for diagnosis of tuberculous meningitis. J. Clin. Microbiol. 38:3150-3155.[Abstract/Free Full Text]
    3. 3
  3. Desjardin, L. E., Y. Chen, M. D. Perkins, L. Teixeira, M. D. Cave, and K. D. Eisenach. 1998. Comparison of the ABI 7700 system (TaqMan) and competitive PCR for quantification of IS6110 DNA in sputum during treatment of tuberculosis. J. Clin. Microbiol. 36:1964-1968.[Abstract/Free Full Text]
    4. 4
  4. Iseman, M. D. 2000. A clinician's guide to tuberculosis. Lippincott Williams & Wilkins, Philadelphia, Pa.
    5. 5
  5. Lang, A. M., J. Feris-Iglesias, C. Pena, J. F. Sanchez, L. Stockman, P. Rys, G. D. Roberts, N. K. Henry, D. H. Persing, and F. R. Cockerill III. 1998. Clinical evaluation of the Gen-Probe amplified direct test for detection of Mycobacterium tuberculosis complex organisms in cerebrospinal fluid. J. Clin. Microbiol. 36:2191-2194.[Abstract/Free Full Text]
    6. 6
  6. Pfyffer, G., P. Kissling, E. Jahn, H. Welscher, M. Salfinger, and R. Weber. 1996. Diagnostic performance of amplified Mycobacterium tuberculosis direct test with cerebrospinal fluid, other nonrespiratory, and respiratory specimens. J. Clin. Microbiol. 34:834-841.[Abstract]
    7. 7
  7. Reischl, U., N. Lehn, H. Wolf, and L. Naumann. 1998. Clinical evaluation of the automated COBAS AMPLICOR MTB assay for testing respiratory and nonrespiratory specimens. J. Clin. Microbiol. 36:2853-2860.[Abstract/Free Full Text]
    8. 8
  8. Sepkowitz, K., J. Raffalli, L. Riley, T. Kiehn, and D. Armstrong. 1995. Tuberculosis in the AIDS era. Clin. Microbiol. Rev. 8:180-199.[Abstract]
    9. 9
  9. Thwaites, G. E., M. Caws, T. T. H. Chau, N. T. Dung, J. I. Campbell, N. H. Phu, T. T. Hien, N. J. White, and J. J. Farrar. 2004. Comparison of conventional bacteriology with nucleic acid amplification (amplified mycobacterium direct test) for diagnosis of tuberculous meningitis before and after inception of antituberculosis chemotherapy. J. Clin. Microbiol. 42:996-1002.[Abstract/Free Full Text]
    10. 10
  10. Thwaites, G. E., T. T. H. Chau, and J. J. Farrar. 2004. Improving the bacteriological diagnosis of tuberculous meningitis. J. Clin. Microbiol. 42:378-379.[Abstract/Free Full Text]
    11. 11
  11. Thwaites, G. E., T. T. H. Chau, K. Stepniewska, N. H. Phu, L. V. Chuong, D. X. Sinh, N. J. White, C. M. Parry, and J. J. Farrar. 2002. Diagnosis of adult tuberculous meningitis by use of clinical and laboratory features. Lancet 360:1287-1292.[CrossRef][Medline]

Journal of Clinical Microbiology, November 2004, p. 5341-5344, Vol. 42, No. 11
0095-1137/04/$08.00+0     DOI: 10.1128/JCM.42.11.5341-5344.2004
Copyright © 2004, American Society for Microbiology. All Rights Reserved.



This article has been cited by other articles:

  • Piersimoni, C., Gherardi, G., Nista, D., Bornigia, S. (2009). Impact of a Chemistry-Based DNA Extraction Method on Performance of a Commercial Amplification Assay for Detection of Mycobacterium tuberculosis Complex. J. Clin. Microbiol. 47: 282-283 [Full Text]  
  • Rock, R. B., Olin, M., Baker, C. A., Molitor, T. W., Peterson, P. K. (2008). Central Nervous System Tuberculosis: Pathogenesis and Clinical Aspects. Clin. Microbiol. Rev. 21: 243-261 [Abstract] [Full Text]  
  • Boudewijns, M., Bakkers, J. M., Sturm, P. D. J., Melchers, W. J. G. (2006). 16S rRNA Gene Sequencing and the Routine Clinical Microbiology Laboratory: a Perfect Marriage?. J. Clin. Microbiol. 44: 3469-3470 [Full Text]  
  • Shah, N. P., Singhal, A., Jain, A., Kumar, P., Uppal, S. S., Srivatsava, M. V. P., Prasad, H. K. (2006). Occurrence of Overlooked Zoonotic Tuberculosis: Detection of Mycobacterium bovis in Human Cerebrospinal Fluid. J. Clin. Microbiol. 44: 1352-1358 [Abstract] [Full Text]  

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Cloud, J. L.
Right arrow Articles by Woods, G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Cloud, J. L.
Right arrow Articles by Woods, G.

Copyright © 2009 by the American Society for Microbiology.   For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»