JOURNAL OF CLINICAL MICROBIOLOGY, June 2005, p. 2563–2566
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
Vol. 43, No. 6
Comparison of the COBAS AMPLICOR MTB and BDProbeTec ET
Assays for Detection of Mycobacterium tuberculosis in
W. H. F. Goessens,1* P. de Man,2J. G. M. Koeleman,2A. Luijendijk,1R. te Witt,2
H. P. Endtz,1and A. van Belkum1
Erasmus MC, University Medical Center Rotterdam, Dept. of Medical Microbiology & Infectious Diseases, Rotterdam,
The Netherlands,1and St. Franciscus Hospital, Dept. of Medical Microbiology, Rotterdam, The Netherlands2
Received 11 June 2004/Returned for modification 17 August 2004/Accepted 13 February 2005
The performances of the BDProbeTec ET (Becton Dickinson) and COBAS AMPLICOR MTB (Roche) were
retrospectively evaluated for detecting Mycobacterium tuberculosis complex in various respiratory specimens.
The BACTEC and MGIT liquid culture system (Becton Dickinson) was used as a reference method. A total of
824 respiratory specimens, comprised of sputa, bronchoalveolar lavage fluid, and bronchial and tracheal
aspirates from 580 patients, were evaluated. Out of 824 clinical specimens, 109 specimens from 43 patients
were culture positive for M. tuberculosis. Of these 109 specimens, 67 were smear positive, 85 were positive by
the COBAS AMPLICOR MTB test, and 94 were positive by the BDProbeTec ET. Of the 715 culture-negative
specimens, 17 were positive by the auramine staining, 11 were positive by the COBAS AMPLICOR MTB test,
and 12 were positive by the BDProbeTec ET. After discrepancy analysis and review of the patients’ clinical data,
130 specimens from 50 patients were considered “true-positive” specimens. This resulted in the following
sensitivities: microscopy, 61.5%; COBAS AMPLICOR MTB test, 78.0%; and BDProbeTec ET, 86.2%. The
specificities of each system, based on the clinical diagnosis, were 99.7% for microscopy, 99.9% for the COBAS
AMPLICOR MTB test, and 99.9% for the BDProbeTec ET. The data presented represent a considerable
number of specimens evaluated with a considerable number of culture- and auramine-positive and culture-
positive and auramine-negative results and therefore give a realistic view of how the data should be interpreted
in a daily routine situation. Specifically, the data with regard to the culture-positive and auramine-negative
specimens are useful, because in a routine situation, auramine-negative specimens are sometimes accepted, on
clinical indications, to be analyzed by an amplification method.
In contrast to general expectations, the incidence of myco-
bacterial disease has significantly increased worldwide since
1990 (7). The technical developments in diagnosing infections
caused by Mycobacterium tuberculosis have increased similarly,
an important change being the replacement of traditional solid
culture media by liquid media and the introduction of molec-
ular amplification techniques, such as PCR, has decreased the
turnaround time even further (1, 22). Many commercial assays
are now available; for instance, the COBAS AMPLICOR MTB
system (Roche, Basel, Switzerland) uses PCR (4, 27), while the
BDProbeTec ET system (14) from Becton Dickinson (Sparks,
Md.) uses the “strand displacement amplification” technique
for detecting M. tuberculosis (3, 5, 6, 8, 11, 23, 24, 26).
Amplification techniques have a good sensitivity for smear-
positive specimens; however, for smear-negative samples, the
reported sensitivity varies considerably (16, 17, 18, 21).
The objective of the present study was to compare the
sensitivities and specificities of the BDProbeTec ET and the
COBAS AMPLICOR MTB properly by evaluating the assays
with the same set of processed specimens. Secondly, as the
smear-negative specimens influence the overall sensitivities of
the amplification methods, the accuracy of the two methods
was studied in a considerable number of culture-positive and
smear-positive specimens as well as in culture-positive but
(Part of this work was presented at the 12th European Con-
gress of Clinical Microbiology and Infectious Diseases, Milan,
Italy, 24 to 27 April 2002.)
MATERIALS AND METHODS
Specimens. Eight hundred twenty-four respiratory specimens (569 sputum
samples or bronchial and tracheal aspirates and 255 bronchoalveolar lavage
specimens) were obtained from 580 patients clinically suspected of suffering from
tuberculosis. All specimens were obtained from two hospitals (the Erasmus
University Medical Center and the St. Franciscus hospital) located in Rotterdam,
Smear examination. Fixed smears of the digested and decontaminated spec-
imens were stained with auramine-rhodamine (Becton Dickinson, Sparks, Md.)
and examined by standard procedures for detection of acid-fast bacilli.
Specimen processing. After digestion, decontamination, and concentration,
the specimens were suspended in phosphate-buffered saline (PBS) and, subse-
quently, one half of the volume was used for culture and one half was stored at
?70°C ready for extraction of the DNA. All specimens were analyzed for the
growth of mycobacteria by inoculation of the concentrated material into the
MGIT or BACTEC system. Inoculation of the MGIT and BACTEC was per-
formed according to the manufacturer’s instructions (Becton Dickinson, Sparks,
Md.). Standardization in both centers was obtained using identical protocols with
regard to specimen processing, auramine staining, and identification of positive
cultures. The only difference between the two centers arises in the volume
applied to resuspend the pellet after the NALC treatment procedure. This part
of the procedure was standardized between the two centers by taking care that
the initial PBS pellet suspension volumes of the pellets were equalized; i.e., in the
Erasmus MC, the pellet suspension was made with up to 2 ml PBS. Otherwise,
results of the amplification methods of the two centers were not comparable. As
* Corresponding author. Mailing address: Erasmus MC, University
Medical Center Rotterdam, Dept. of Medical Microbiology & Infec-
tious Diseases, Dr. Molewaterplein 40, 3015 GD Rotterdam, The
Netherlands. Phone: 31 (0) 10 463-6171. Fax: 31 (0) 10 463 3875.
quality controls, positive and negative amplification controls were included in
each COBAS AMPLICOR or BDProbeTec ET run. The COBAS AMPLICOR
and BDProbeTec ET procedures were followed in accordance with the manu-
facturer’s instructions (11, 12, 15)
AccuProbe culture identification tests (AccuProbe; Gen-Probe, Inc., San Di-
ego, Calif.) were used for the identification of the isolates.
Discrepancy analysis. A positive culture on Lowenstein-Jensen slants and/or
in BACTEC or MGIT liquid culture systems which was also confirmed by acid-
fast bacilli and AccuProbe was regarded as a true-positive specimen.
Any positive result obtained using the COBAS AMPLICOR MTB or
BDProbeTec ET system that was not confirmed by a positive culture (solid or
liquid medium) was regarded as a false-positive amplification result.
Positive results obtained using either COBAS AMPLICOR MTB or
BDProbeTec ET system that were also confirmed by positive culture (solid or
liquid medium) were regarded as true-positive amplification results.
Any specimen found to be positive with the COBAS AMPLICOR MTB
system as well as with the BDProbeTec ET system but negative with solid or
liquid medium was regarded as a true-positive result for diagnosis but a false-
negative result for tuberculosis culture with solid or liquid medium.
Culture-negative specimens that were positive with either the COBAS
AMPLICOR MTB or the BDProbeTec ET system were regarded as M. tuber-
culosis positive when patients were clinically suspected of suffering from pulmo-
nary tuberculosis and responding to antimicrobial therapy. Patients were also
considered positive if a tuberculosis culture-positive specimen had been obtained
(solid or liquid medium) in the last 6 months prior to the study.
The results of this evaluation are shown in Table 1. One
hundred nine out of 824 specimens were culture positive for
M. tuberculosis. Of these 109 culture positives, the COBAS
AMPLICOR MTB system detected 85, and another 11 speci-
mens were positive by this method in the M. tuberculosis cul-
ture-negative group. These results indicate that the indepen-
dent sensitivity and specificity for the COBAS AMPLICOR
MTB assay were 78 and 98.5%, respectively. Similarly, the
BDProbeTec ET system detected 94 out of the 109 culture-
positive specimens, with 12 positives among the culture-nega-
tive specimens. The sensitivity and specificity scores of the
BDProbeTec ET were therefore calculated to be 86.2% and
98.3%, respectively. Interestingly, 17 culture-negative speci-
mens were positive by at least one of the amplification systems
used and 6 out of 17 of these negatives were actually positive
by both amplification reactions. The remaining 11 discrepant
results were resolved by studying the medical history of the
patients from whom the 11 discrepant results came. These pa-
tients either had been clinically suspected of suffering from
pulmonary tuberculosis, had responded to antimicrobial ther-
apy, or had been pretreated for M. tuberculosis infection dur-
ing the last 6 months. Consequently, 9 out of 11 COBAS
AMPLICOR MTB or BDProbeTec ET positives with neg-
ative culture results were regarded as true positives, giving
rise to identical specificities for both assays, i.e., 99.9%.
Thus, there appeared to be only one false-positive result
using each amplification system for the 824 specimens tested.
The sensitivities for culture-positive, auramine-negative, and
auramine-positive specimens differ. Among the auramine-pos-
itive specimens, BDProbeTec ET detected 67/67 (100%) and
the COBAS AMPLICOR MTB detected 64/67 (95.5%). For
the auramine-negative specimens, a small difference was ob-
served, as it was found that the BDProbeTec ET system de-
tected 27/42 (64.3%) and the COBAS AMPLICOR MTB de-
tected 21/42 (50%) auramine-negative specimens.
The primary objective of this study was to compare two
amplification techniques in their ability to detect M. tubercu-
losis in respiratory specimens. Culture of M. tuberculosis in a
liquid medium and auramine staining were the reference
At the Erasmus MC site, the COBAS AMPLICOR MTB
technique is routinely applied together with the MGIT as a
liquid culture technique. At the other collaborating center, the
BDProbeTec ET technique and the BACTEC liquid culture
technique are applied together in the routine setting. After
specimens had been processed for culture and microscopy,
subsamples of the specimens were exchanged between the par-
ticipating centers to be analyzed with the respective amplifica-
tion methods (2).
Although the specimen volume was corrected, different re-
sults were obtained using the two amplification systems, as
the COBAS AMPLICOR MTB system detected 85/109 (78%)
and BDProbeTec ET system detected 94/109 (86.2%) culture-
positive specimens. The majority of the differences found
between the two systems came from the auramine-negative
specimens. In 42 auramine-negative but culture-positive spec-
imens, BDProbeTec ET detected 27 (64.3%) and the COBAS
AMPLICOR MTB detected 21 (50%) positive specimens. The
limited value of amplification assays in detecting true positives
in auramine-negative specimens has been reported earlier by
Piersimoni et al. (19).
Our results were comparable with the results obtained by
Ichayama et al. (9), who reported sensitivities of 89.5% and
94.7% for COBAS AMPLICOR MTB and BDProbeTec ET,
respectively. Similar results were obtained by Kim et al. (12);
however, interpretation of these data is difficult, as only a
limited number (n ? 26) of positive specimens were studied.
Iinuma (10) did not find significant differences between the
two amplification methods, but this is due to the fact that the
specimens evaluated were almost all auramine positive. From
our data, it is shown that there is no difference between the two
amplification methods for the auramine-positive specimens.
The differences between the amplification methods are ob-
served with the auramine-negative specimens.
The COBAS AMPLICOR MTB and BDProbeTec ET sys-
tems detected additional positive specimens, though after
TABLE 1. Comparison of culture results for 824 respiratory
specimens with those obtained by staining, BDProbeTec ET, and
COBAS AMPLICOR MTB systems for detection of M. tuberculosis
Assay and result
No. of cultures with
(n ? 109)
(n ? 715)
COBAS AMPLICOR MTB
11 78.0 98.5
2564GOESSENS ET AL.J. CLIN. MICROBIOL.
studying the medical history of the patients involved, we found
that most of these additionally positive specimens came from
patients who were undergoing treatment or who had been
previously treated for M. tuberculosis during the last 6 months.
Hellyer et al., Levee et al., and Thomsen et al. (8, 13, 25)
previously demonstrated that M. tuberculosis DNA could be
detected by strand displacement amplification and PCR in
frozen sputum samples more than 12 months after the initia-
tion of treatment and more than 6 months after positive-to-
negative culture conversion.
The differences found between the two amplification sys-
tems could be explained by differences in the extraction meth-
ods used or by a reduction in the amount of inhibitory factors.
The difference in sensitivity cannot be explained by differences
in the volume used in the two amplification reactions. In the
COBAS AMPLICOR MTB system, 25 ?l of the primary ex-
tract is actually used in the amplification reaction. In the
BDProbeTec ET assay, 13.75 ?l is used. However, when the
input volume is reduced, the input of inhibitory substances is
also reduced, and the difference in sensitivities is therefore
probably due to a better extraction method in combination
with a reduction in inhibition, leading to optimized amplifica-
tion. This presumably explains the absence of inhibited sam-
ples for the BDProbeTec ET assay in our study. In the COBAS
AMPLICOR MTB system, seven specimens (0.85%) were re-
ported to be inhibited, which leads to inconclusive results.
Inhibition rates for the COBAS AMPLICOR MTB varying
from 0.3% to 3.9% for respiratory and nonrespiratory spec-
imens have been reported by others (20, 21, 25). In contrast
to these findings, Kim et al. (12) reported an inhibition
rate of 0.7% for BDProbeTec ET and of 0% for COBAS
AMPLICOR. As the inhibition rates vary considerably, we
suggest that the differences found between the two methods
are probably due to a better extraction protocol applied by the
BDProbeTec ET technique. Auramine-negative specimens
have a low number of bacteria, and with these specimens,
differences between results of the two methods are observed.
Probably, due to the mechanical disruption applied by the
BDProbeTec ET technique, more specific DNA of the organ-
isms is extracted and becomes available for amplification.
At present, liquid culture techniques still appear to be
superior to the amplification techniques in the diagnosis of
M. tuberculosis infection. The difference in sensitivities be-
tween culture and molecular (standard PCR) testing is most
probably due to the differences in input volume; culture is bet-
ter due to a larger inoculum size. In our study, the advantage
of the molecular techniques is lost for the auramine-negative
group, i.e., the specimens with a low number of acid-fast bacilli.
From our results, the COBAS AMPLICOR MTB and
BDProbeTec ET techniques have an added value for speci-
mens that are also found to be auramine positive. In this re-
spect, the COBAS AMPLICOR MTB and BDProbeTec ET
amplification systems easily discriminate between M. tubercu-
losis and non-M. tuberculosis isolates, giving relevant informa-
tion to the clinician with respect to diagnosis and optimal
treatment of a particular patient. Both COBAS AMPLICOR
MTB and BDProbeTec ET systems have an acceptable sensi-
tivity and an excellent specificity. A difference in sensitivities
between the results obtained for the present study and the
results presented in previous publications may be attributed to
the use of different reference methods, the use of different pa-
tient specimens (i.e., tissue specimens), or the inclusion or ex-
clusion of auramine-negative specimens.
In conclusion, the sensitivities of the BDProbeTec ET sys-
tem and the COBAS AMPLICOR MTB appear adequate for
the analysis of auramine-positive samples. For auramine-neg-
ative samples, however, the number of false-negative results is
still too high. Our results confirm the prevailing opinion that
amplification techniques should currently serve only as an ad-
junct to culture and certainly not as a replacement.
We thank John Hays for critically reading and improving the manu-
This work was supported by a grant from Roche Diagnostics and
Becton Dickinson and Co.
1. American Thoracic Society. 1997. Rapid diagnostic tests for tuberculosis.
What is the appropriate use? Am. J. Respir. Crit. Care Med. 155:1804–1814.
2. American Thoracic Society. 2000. Diagnostic standards and classification of
tuberculosis in adults and children. Am. J. Respir. Crit. Care Med. 161:
3. Barrett, A., J. G. Magee, and R. Freeman. 2002. An evaluation of the BD
ProbeTec ET system for the direct detection of Mycobacterium tuberculosis
in respiratory samples. J. Med. Microbiol. 51:895–898.
4. Bergmann, J. S., and G. L. Woods. 1996. Clinical evaluation of the Roche
AMPLICOR PCR Mycobacterium tuberculosis test for detection of M. tuber-
culosis in respiratory specimens. J. Clin. Microbiol. 34:1083–1085.
5. Bergmann, J. S., W. E. Keating, and G. L. Woods. 2000. Clinical evaluation
of the BDProbeTec ET system for rapid detection of Mycobacterium tuber-
culosis. J. Clin. Microbiol. 38:863–865.
6. Down, J. A., M. A. O’Connell, M. S. Dey, A. H. Walters, D. R. Howard, M. C.
Little, W. E. Keating, P. Zwadyk, Jr., P. D. Haaland, D. A. McLaurin III, and
G. Cole. 1996. Detection of Mycobacterium tuberculosis in respiratory spec-
imens by strand displacement amplification of DNA. J. Clin. Microbiol. 34:
7. Dye, C., S. Scheele, P. Dolin, V. Pathania, and M. C. Raviglione. 1999.
Global burden of tuberculosis: estimated incidence, prevalence, and mortal-
ity by country. Consensus statement: W.H.O. global surveillance and moni-
toring project. JAMA. 282:677–686.
8. Hellyer, T. J., T. W. Fletcher, J. H. Bates, W. W. Stead, G. L. Templeton,
M. D. Cave, and K. D. Eisenach. 1996. Strand displacement amplification
and the polymerase chain reaction for monitoring response to treatment in
patients with pulmonary tuberculosis. J. Infect. Dis. 173:934–941.
9. Ichiyama, S., Y. Ito, F. Sugiura, Y. Iinuma, Y. Sadaaki, M. Shimojima, Y.
Hasegawa, K. Shimokata, and N. Nakashima. 1997. Diagnostic value of the
strand displacement amplification method compared to those of Roche
Amplicor PCR and culture for detecting mycobacteria in sputum samples.
J. Clin. Microbiol. 35:3082–3085.
10. Iinuma, Y., K. Senda, N. Fujihara, T. Saito, S. Takakura, M. Shimojima, T.
Kudo, and S. Ichiyama. 2003. Comparison of the BDProbeTec ET system
with the COBAS AMPLICOR PCR for direct detection of Mycobacterium
tuberculosis in respiratory samples. Eur. J. Clin. Microbiol. Infect. Dis. 22:
11. Johansen, I. S., V. O. Thomsen, A. Johansen, P. Andersen, and B. Lundgren.
2002. Evaluation of a new commercial assay for diagnosis of pulmonary and
nonpulmonary tuberculosis. Eur. J. Clin. Microbiol. Infect. Dis. 21:455–460.
12. Kim, S. Y., Y. J. Park, S. J. Kang, B. K. Kim, and C. S. Kang. 2004.
Comparison of the BDProbeTec ET system with the Roche COBAS
AMPLICOR system for detection of Mycobacterium tuberculosis complex in
the respiratory and pleural fluid specimens. Diagn. Microbiol. Infect. Dis. 49:
13. Levee, G., P. Glaziou, B. Gicquel, and S. Chanteau. 1994. Follow-up of
tuberculosis patients undergoing standard anti-tuberculosis chemotherapy
by using a polymerase chain reaction. Res. Microbiol. 45:5–8.
14. Little, M. C., J. Andrews, R. Moore, S. Bustos, L. Jones, C. Embres, G.
Durmowicz, J. Harris, D. Berger, K. Yanson, C. Rostkowski, D. Yursis, J.
Price, T. Fort, A. Wlaters, M. Collis, O. Liorin, J. Wood, F. Failing, C.
O’Keefe, B. Scrivens, B. Pope, T. Hansen, K. Williams, and M. Boenisch.
1999. Strand displacement amplification and homogeneous real-time detec-
tion incorporated in a second generation DNA probe system, BDProbeTec
ET. Clin. Chem. 45:777–784.
15. Mazzarelli, G., L. Rindi, P. Piccoli, C. Scarparo, C. Garzelli, and E. Tortoli.
2003. Evaluation of the BDProbeTec ET system for direct detection of
Mycobacterium tuberculosis in pulmonary and extrapulmonary samples: a
multicenter study. J. Clin. Microbiol. 41:1779–1782.
VOL. 43, 2005 DETECTION OF M. TUBERCULOSIS IN RESPIRATORY SPECIMENS2565
16. Oh, E.-J., Y.-J. Park, C. L. Chang, B. K. Kim, and S. M. Kim. 2001. Download full-text
Improved detection and differentiation of mycobacteria with combination of
Mycobacterium Growth Indicator Tube and Roche COBAS AMPLICOR
System in conjunction with duplex PCR. J. Microbiol Methods 46:29–36.
17. Pfyffer, G. E., P. Kissling, E. M. Jahn, H. M. Welscher, M. Salfinger, R.
Weber. 1996. Diagnostic performance of Amplified Mycobacterium tubercu-
losis Direct Test with cerebrospinal fluid, other nonrespiratory, and respira-
tory specimens. J. Clin. Microbiol. 34:834–841.
18. Piersimoni, C., A. Callegaro, C. Scarparo, V. Penati, D. Nista, S. Bornigia,
C. Lacchini, M. Scagnelli, G. Santini, and G. de Sio. 1998. Comparative
evaluation of the new Gen-Probe Mycobacterium tuberculosis Amplified Di-
rect Test and the semiautomated Abbott LCx Mycobacterium tuberculosis
assay for direct detection of Mycobacterium tuberculosis complex in respira-
tory and extrapulmonary specimens. J. Clin. Microbiol. 36:3601–3604.
19. Piersimoni, C., and C. Scarparo. 2003. Relevance of commercial amplifica-
tion methods for direct detection of Mycobacterium tuberculosis complex in
clinical samples. J. Clin. Microbiol. 41:5355–5365.
20. Reischl, U., N. Lehn, H. Wolf, and L. Naumann. 1998. Clinical evaluation of
the automated COBAS AMPLICOR MTB assay for testing respiratory and
non-respiratory specimens. J. Clin. Microbiol. 36:2853–2860.
21. Scarparo, C., P. Piccoli, A. Rigon, G. Ruggiero, M. Scagnelli, and C. Pier-
simoni. 2000. Comparison of enhanced Mycobacterium tuberculosis amplified
direct test with COBAS AMPLICOR Mycobacterium tuberculosis assay for
direct detection of Mycobacterium tuberculosis complex in respiratory and
extrapulmonary specimens. J. Clin. Microbiol. 38:1559–1562.
22. Schluger, N. W. 2001. Changing approaches to the diagnosis of tuberculosis.
Am. J. Respir. Crit. Care Med. 164:2020–2024.
23. Spargo, C. A., P. D. Haaland, S. R. Jurgensen, D. D. Shank, and G. T.
Walker. 1993. Chemiluminescent detection of strand displacement amplified
DNA from species comprising the Mycobacterium tuberculosis complex. Mol.
Cell. Probes 7:395–404.
24. Thierry, D., A. Brisson-Noel, V. Vincent-Levy-Frebault, S. Nguyen, J.-L.
Guesdon, and B. Gicquel. 1990. Characterization of a Mycobacterium tuber-
culosis insertion sequence, IS6110, and its application in diagnosis. J. Clin.
25. Thomsen, V. O., A. Kok-Jensen, M. Buser, S. Philippi-Schulz, and H. J.
Burkardt. 1999. Monitoring treatment of patients with pulmonary tubercu-
losis: can PCR be applied? J. Clin. Microbiol. 37:3601–3607.
26. Walker, G. T., J. G. Nadeau, P. A. Spears, J. L. Schram, C. M. Nycz, and
D. D. Shank. 1994. Multiplex strand displacement amplification (SDA) and
detection of DNA sequences from Mycobacterium tuberculosis and other
mycobacteria. Nucleic Acids Res. 22:2670–2677.
27. Wobeser, W. L., M. Krajden, J. Conly, H. Simpson, B. Yim, M. D’Costa, M.
Fuksa, C. Hian-Cheong, M. Patterson, A. Phillips, R. Bannatyne, A. Had-
dad, J. L. Brunton, and S. Krajden. 1996. Evaluation of Roche AMPLICOR
PCR assay for Mycobacterium tuberculosis. J. Clin. Microbiol. 34:134–139.
2566GOESSENS ET AL. J. CLIN. MICROBIOL.