See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/24146086

Detecting Mycobacterium tuberculosis in Bactec MGIT 960 Cultures by

Inhouse IS6110-based PCR Assay in Routine Clinical Practice

Article  in  Journal of the Formosan Medical Association · March 2009

DOI: 10.1016/S0929-6646(09)60042-5 · Source: PubMed

CITATIONS READS

25 292

4 authors, including:

Jun-Ren Sun Shih yi Lee

National Defense Medical Center Taipei Veterans General Hospital
50 PUBLICATIONS   552 CITATIONS    30 PUBLICATIONS   677 CITATIONS   

SEE PROFILE SEE PROFILE

Cherng-Lih Perng
Tri-Service General Hospital
44 PUBLICATIONS   603 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Acinetobacter baumannii View project

Other project View project

All content following this page was uploaded by Cherng-Lih Perng on 05 November 2014.

The user has requested enhancement of the downloaded file.

 ORIGINAL ARTICLE

Detecting Mycobacterium tuberculosis in

Bactec MGIT 960 Cultures by Inhouse
IS6110-based PCR Assay in Routine
Clinical Practice
Jun-Ren Sun,1 Shih-Yi Lee,1,2 Cherng-Lih Perng,1 Jang-Jih Lu1,3*

Background/Purpose: Diagnosis of tuberculosis is challenging because the current methods are time-
consuming and laborious. We have developed a method combining the Bactec MGIT 960 rapid culture
system with the IS6110-based PCR for rapid diagnosis of tuberculosis.
Methods: A total of 1745 samples from 712 patients treated at the Tri-Service General Hospital, Taipei,
Taiwan between June and August 2005 were tested. An aliquot of positive Bactec MGIT 960 culture fluids
was Kinyoun stained, and the samples positive for Kinyoun staining were directly assayed by the IS6110-
based PCR. The same samples were also examined by the conventional methods for identification of
Mycobacterium tuberculosis complex.
Results: One hundred and four samples from 62 patients were positive according to the Bactec MGIT 960
system. Among these, 59 (56.7%) were positive and 45 (43.3%) were negative according to the IS6110-
based PCR. Compared with the conventional identification methods, the IS6110-based PCR assay correctly
identified all 59 M. tuberculosis isolates and gave negative results for all nontuberculosis mycobacteria. The
mean turnaround time for M. tuberculosis identification by this combined method was 6.41 days for
smear-positive and 14.33 days for smear-negative specimens. The sensitivity of the IS6110-based PCR assay
was determined to be 5 cells or 0.1 pg of mycobacterial DNA.
Conclusion: The combined use of the automated Bactec MGIT 960 system and the IS6110-based PCR assay
is sensitive and rapid for the detection of M. tuberculosis complex, and we recommend that this method be
used routinely for identification of mycobacteria in clinical laboratories. [J Formos Med Assoc 2009;108(2):
119–125]

Key Words: Bactec MGIT 960 system, IS6110-based PCR assay, Mycobacterium tuberculosis, rapid identification

Tuberculosis continues to have a high fatality rate 7H11 medium are time-consuming with a turn-
worldwide, and its diagnosis continues to depend around time of 3–8 weeks, and the sensitivity for
on microscopy and culture. Microscopy follow- detection of mycobacteria with solid culture
ing Ziehl-Neelsen or auramine-rhodamine stain- media is only 10–15%.2 However, rapid and ac-
ing has a sensitivity ranging from 50% to 80% curate identification of Mycobacterium tuberculosis
as compared to culture.1 Cultures on egg-based (MTB) in clinical specimens is crucial in order to
Lowenstein-Jensen or agar-based Middlebrook prevent transmission of this organism.

©2009 Elsevier & Formosan Medical Association

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1
Division of Clinical Pathology, Department of Pathology, Tri-Service General Hospital and National Defense Medical
Center, 2Departments of Microbiology and Pathology and Laboratory Medicine, Taipei Veterans General Hospital,
Taipei, and 3Department of Laboratory Medicine, China Medical University Hospital, Taichung, Taiwan.

Received: April 28, 2008 *Correspondence to: Dr Jang-Jih Lu, Department of Laboratory Medicine, China Medical
Revised: June 23, 2008 University Hospital, 2 Yuh-Der Road, Taichung 404, Taiwan.
Accepted: July 8, 2008 E-mail: janglu45@gmail.com

J Formos Med Assoc | 2009 • Vol 108 • No 2 119

J.R. Sun, et al
The use of liquid culture media has improved
the sensitivity in the detection of mycobacteria.
These methods include the radiometric Bactec
460TB system (Becton Dickinson Microbiology Sys-
tems, Sparks, MD, USA)2,3 and the nonradiometric
Septi-Chek AFB system (Becton Dickinson)4,5 that
are used in most clinical laboratories. The Bactec
MGIT 960 system (Becton Dickinson, Heidelberg,
Germany) is a novel liquid culture system for
mycobacteria; it is fully automated and is nonra-
diometric and noninvasive. Previous studies have
demonstrated that the Bactec MGIT 960 system
is sensitive and shortens the turnaround time by
2–27 days.6–11
However, identification of M. tuberculosis by
conventional biochemical tests still has a very long
turnaround time of 3–5 weeks, leading to signif-
icant delays in diagnosis. Amplification techniques
such as polymerase chain reaction (PCR) have
been shown to be effective in the identification of
mycobacteria from clinical specimens.12–16 The in-
sertion sequence IS6110 is usually found in species
of the M. tuberculosis complex (MTBC) including
M. tuberculosis, M. bovis, M. africanum, M. microti
and M. bovis BCG.17,18 Since IS6110 is present in
M. tuberculosis in multiple copies, it is an excel-
lent target for detection of the organism by DNA
amplification.19,20 In this study, we evaluated its
efficacy in the identification of mycobacteria when
combined with the IS6110-based PCR.
Methods
Specimen processing and identification of
mycobacteria by conventional methods
A total of 1745 clinical specimens including 1557
sputum samples, 123 pleural fluid samples, 14
bronchial brushings, 13 abscess samples, 10 cere-
brospinal fluid samples, 10 peritoneal fluid sam-
ples, 9 urine samples, 5 ascites fluid samples, and
4 tissue samples were collected from 712 patients
suspected of having mycobacterial infections be-
tween June and August 2005. The specimens
were processed within 24 hours of collection as
described previously.6 All specimens were first
120
decontaminated with an equal volume of N-
acetyl-L-cysteine (NALC)–NaOH (final concentra-
tion: 2% NaOH and 0.5% NALC) for 15 minutes
at room temperature and then neutralized with
sterile 0.067 M phosphate buffer (pH 6.8). After
centrifugation at 3000g for 15 minutes, the sedi-
ment was resuspended in 2.0 mL of phosphate
buffer. An aliquot of the suspension was stained
with auramine-rhodamine stain and Kinyoun
stain to detect acid-fast bacilli (AFB); 0.5 mL of the
same suspension was inoculated into a Bactec
MGIT 960 tube, and another 0.2 mL was inoculated
onto one Lowenstein-Jensen (LJ) slant (Becton
Dickinson, Heidelberg, Germany). The cultures
were incubated for 8 weeks. The LJ medium was
examined on alternate days from day 2 to day 7
and once a week from day 8 to day 56 according
to established laboratory procedures.6
Bactec MGIT 960 tubes that yielded a positive
culture signal were considered positive for my-
cobacteria only after confirmation by Kinyoun
stain. An aliquot of the positive Bactec MGIT 960
culture was inoculated onto two LJ slants for fur-
ther identification. Bactec MGIT 960 tubes with a
positive culture signal but failed to reveal AFB in
the smear were considered contaminated and elim-
inated from further study. If mycobacteria were
detected in Bactec MGIT 960 tubes or on LJ me-
dium, they were identified by biochemical tests.4
The bacteria were grown on LJ slants and exam-
ined for growth rate, colony morphology and pig-
mentation. Identification of MTB was performed
by nitrate reduction and niacin test. Nontuber-
culous mycobacterium (NTM) identification to
the species level was achieved by the 16S rDNA
sequencing technique21,22 and a battery of bio-
chemical reactions, including Tween 80 hydrolysis,
nitrate reduction, arylsulfatase, and semiquanti-
tative catalase reaction.4
Identification of MTB from positive MGIT by
IS6110-based PCR
After the presence of AFB was confirmed by
Kinyoun staining in positive MGIT cultures, the
bacteria in the tube were examined by IS6110-
based PCR. The positive MGIT tube was mixed
J Formos Med Assoc | 2009 • Vol 108 • No 2

gently and precipitated spontaneously for 10 min-
utes. Then, 50 μL of the culture fluid from the su-
pernatant of the tube were taken and incubated in
400 μL TE buffer (10 mM Tris-HCl, 1 mM EDTA,
pH 7.0) for 30 minutes at 80°C. DNA was then
extracted as previously described.23
The DNA precipitate was dissolved in 50 μL of
TE buffer and stored frozen at −20°C until PCR
was performed. Then, 5 μL of the extracted DNA
was added to 45 μL of reagent mix, which con-
tained 5 μM of TB762F primer (5⬘-CTC GTC CAG
CGC CGC TTC GG-3⬘) and TB884R primer (5⬘-
CCT GCG AGC GTA GGC GTC GG-3⬘), 250 μM
of each dNTP, 2.0 mM MgCl2 in 1 × EXT buffer
(50 mM Tris-HCl, 15 mM (NH4)2SO4, 1.5 mM
MgCl2, 0.1% Triton X-100, pH 9.0), and 0.5 U of
DyNAzyme EXT (Finnzymes Oy, Espoo, Finland).
The amplification mix was placed in a ther-
mal cycler (GeneAmp 2400; Applied Biosystems,
Foster City, CA, USA) and run for 1 cycle at 96°C
for 2 minutes; 35 cycles at 96°C for 15 seconds
and 60°C for 1 minute; and 1 cycle at 72°C for
7 minutes. A plasmid containing primer binding
sites was used as the internal control to yield a
600-bp DNA band in each PCR reaction.
PCR products were analyzed by electrophoresis
on 2% agarose gel or 6% PAGE followed by stain-
ing with ethidium bromide. The amplicon specific
for MTBC was 123 bp.16,19 The quality of the am-
plification was monitored by the simultaneous
testing of a blank and a positive control sample in
each run. The blank control samples consisted of
water. The positive TB control samples consisted
of mycobacterial strain, M. tuberculosis H37Rv.
Sensitivity of IS6110-based PCR
The lowest limit of detection of bacterial cells or
purified M. tuberculosis DNA by PCR was deter-
mined using strain H37Rv. One hundred nano-
gram of DNA was 10-fold serially diluted in sterile
distilled water. Bacterial suspensions containing
1 × 107 cells/mL were similarly diluted in normal
saline. DNA isolated from 100 μL of each diluted
bacterial suspension was added directly to the
PCR mixture. Bacterial number was determined
by plating 0.1 mL of each dilution onto 7H11
J Formos Med Assoc | 2009 • Vol 108 • No 2
Combination of MGIT and PCR for TB diagnosis
agar plates. All PCR reactions were performed in
triplicate.
Identification by 16S rDNA sequencing
An 804-bp fragment of the 16S rDNA gene of each
isolate was amplified by PCR in a total reaction
volume of 50 μL containing 250 μM of each
dNTP, 2.0 mM MgCl2 in 1 × EXT buffer and 0.5 U
of DyNAzyme EXT (Finnzymes Oy). Thermal cy-
cling consisted of an initial denaturation at 95°C
for 5 minutes followed by 40 cycles of denaturation
at 95°C for 30 seconds, primer annealing at 55°C
for 30 seconds, and extension at 72°C for 60 sec-
onds with a single final extension at 72°C for 10
minutes. The broad-range primers 16S-5f (5⬘-GAA
GAG TTT GAT CMT GGC TC-3⬘) and 16S-809r
(5⬘-GCG TGG ACT ACC AGG GTA TC-3⬘) were
used for the PCR.22 A plasmid containing identical
or near-identical primer binding sites was used as
the internal control as described previously.21,22
In order to control for the presence of contam-
inating nucleic acids, controls containing water in
place of template DNA were run in parallel. The
PCR product was purified using the Virogene PCR
clean-up kit (Virogen, Taiwan) according to the
manufacturer’s instructions. Nucleotide sequences
were determined by using an ABI Prism 377 genetic
analyzer (Applied Biosystems) and the BigDye ter-
minator cycle sequencing ready reaction kit (ver-
sion 3.1; Applied Biosystems). The sequencing
reaction included 2 μL of Premix, 5 pmol of se-
quencing primer, and 0.2 μg of the PCR product
in a total volume of 10 μL. The primer 16S-531r
(5’-TAC CGC GGC TGC TGG CA-3’) was used
to sequence the hypervariable region of the 16S
rRNA gene. Identification of NTM isolates was
performed by comparing their sequences against
the sequences in the GenBank database.
Results
Sensitivity of IS6110-based PCR
The criterion for a positive PCR assay specific for
MTB was a visible amplified product of 123 bp in
the gel after electrophoresis. A 600 bp amplification
121
J.R. Sun, et al

product of the internal control was detected for
all PCR reactions and negative controls (Figure).
The IS6110-based PCR assay had a detection limit
of 5 cells or 0.1 pg of mycobacterial DNA.
Turnaround time for identification of
mycobacteria by traditional biochemical tests
Among the 1745 clinical specimens, 243 MGIT
tubes yielded a positive culture signal. Of these
M 1 2 3 4 M
600
500
400
300
200
100
Figure. Gel electrophoresis analysis of IS6110 PCR products.
Lanes: M = 100 bp marker; 1 = positive specimen with a 123-
bp IS6110 fragment and a 600-bp amplicon of the internal
control; 2 = negative specimen with a 600-bp amplicon of
the internal control; 3 = blank control; 4 = positive control.
MGIT positive tubes, 139 were considered to be
contaminated and eliminated. The remaining 104
MGIT positive tubes were confirmed to contain
AFB by Kinyoun staining of culture fluid smears.
Times to detection of mycobacterial growth by
the Bactec MGIT 960 system are summarized in
Table 1. The mean turnaround time for identifi-
cation of MTB was 6.41 days for smear-positive
and 14.33 days for smear-negative specimens,
and those for rapid growing NTM were 5 days for
smear-positive and 8 days for smear-negative
specimens. The mean turnaround time for iden-
tification of slow growing NTM was 15.83 days
for smear-positive specimens and 18.75 days for
smear-negative specimens. The overall mean turn-
around time for identification of the 104 MGIT
positive specimens was 11.67 days.
Sensitivity and specificity of IS6110-based
PCR for rapid identification of MTB
Results of the 104 culture positive isolates are
shown in Table 2. Among the 104 MGIT positive
samples, 59 (56.73%) were positive and 45
(43.27%) were negative according to the IS6110-
based PCR. The 104 MGIT positive tubes were
from 62 different patients and the 59 IS6110-
based PCR positive samples were from 28 pa-
tients. This result was identical to that of the
conventional biochemical tests. With the 16S
rDNA sequencing identification, the 45 NTM
isolates were found to include 14 M. chelonae,

Table 1. Culture results and turnaround time (days) of the Bactec MGIT 960 system
Positive mycobacterial Mean time in days
Species Smear result*
cultures (n) (range)

MTB AFB positive 27 6.41 (4–13)

AFB negative 27 14.33 (7–37)
Rapid-growing NTM AFB positive 1 5
AFB negative 21 8 (3–25)
Slow-growing NTM AFB positive 6 15.83 (5–37)
AFB negative 16 18.75 (6–35)
Culture only† 6 14.33 (10–19)
Total 104 11.67 (3–37)
*Smears were stained with Kinyoun stain after decontamination; †no staining was done. MTB = Mycobacterium tuberculosis; NTM =
nontuberculous mycobacterium; AFB = acid-fast bacilli.

122 J Formos Med Assoc | 2009 • Vol 108 • No 2


Table 2. Results of IS6110-based PCR from MGIT
cultures
IS6110 PCR result
Mycobacteria species Positive Negative
samples (n) samples (n)
M. tuberculosis complex 59
M. chelonae group 14
M. fortuitum group 8
M. avium complex 10
M. gordornae 4
M. kansasii 7 The contamination rate in this study was 57.5%
M. szulgai 1 (139/243), which is similar to the 48.7–80.6%
M. terrae complex 1 reported by Williams-Bouyer et al,28 Scarparo
Total 59 45
8 M. fortuitum, 10 M. avium complex, 4 M. gordor-
nae, 7 M. kansasii, 1 M. szulgai, and 1 M. terrae
complex. Comparison of IS6110-based PCR with
conventional biochemical tests in the 104 MGIT
positive tubes found no significant difference
(p = 1.0, Fisher’s exact test).
Discussion
Rapid diagnosis of mycobacterial infections is criti-
cal; therefore, attempts to shorten the time required
for detection and identification of mycobacteria
deserve attention. The guidelines of the Center for
Disease Control and Prevention (CDC) recom-
mend that the turnaround time for detection and
identification should be 2–3 weeks and 2–4 weeks
for susceptibility testing after receipt of the speci-
men.9,24,25 Previous studies have shown that the
fully automated MGIT 960 is better in sensitivity
and turnaround time than other culture sys-
tems.6–11 The higher sensitivity and shorter detec-
tion time with the MGIT system are due to the use
of an oxygen-sensitive fluorescent compound. With
regard to the turnaround time, the mean detec-
tion time was significantly shorter for methods
that used a liquid medium than the LJ medium.7
In our study, the mean turnaround time for de-
tection of MTB in smear-positive specimens was
6.41 days (range, 4–13 days). This is a significant
J Formos Med Assoc | 2009 • Vol 108 • No 2
Combination of MGIT and PCR for TB diagnosis
improvement from the 9.9–12.6 days reported
by Hanna et al,26 Tortoli et al,6 Pfyffer et al27 and
Somoskovi et al.8,10 Our turnaround time for de-
tection of MTB in smear-negative specimens was
14.33 days (range, 7–37 days), slightly shorter than
the 15.8–20.3 days reported by Hanna et al,26
Tortoli et al,6 Pfyffer et al27 and Somoskovi et al.8,10
The contamination rates of tuberculosis culture
for both systems were much higher than that
of solid medium systems in other studies.28–30
et al,29 Hillemann et al30 and Kontos et al.31 To
avoid a high contamination rate, the use of decon-
tamination procedures with an increased concen-
tration of NaOH (3% or 4%), decreased bacterial
contamination rates.30 Despite this, it is likely
that a more appropriate solution preventing any
decontamination-related mycobacterial damage
should rely on the revision of the antibiotic sup-
plement, especially by adding new drugs able to
control Gram-positive overgrowth.29
Rapid detection by the MGIT system is of lim-
ited benefit if it is not complemented by an accurate
and rapid identification method. Rapid identifica-
tion is best achieved by molecular biological meth-
ods, but only a limited number of evaluations have
been reported on the identification of MTBC from
MGIT cultures.14–16,32–34 Recently, three commer-
cial PCR based kits, the rRNA amplification-based
Gen-Probe Amplified Mycobacterium Tuberculosis
Direct Test system (Accuprobe; Gen-Probe Inc.,
San Diego, CA, USA),32 the COBAS AMPLICOR
MTB PCR (COBAS MTB) assay (Roche Diagnostics,
Mannheim, Germany),35 and the GenoType My-
cobacterium assay (Hain Diagnostika, Nehren,
Germany),14 have provided the speed and relia-
bility for direct detection and identification of
mycobacteria. The disadvantages of these kits,
however, are their high costs and the requirement
of additional equipment.16
Another culture confirmation test is the
CapiliaTB (MPB64-ICA; TAUNS, Numazu, Japan)
for MTBC which uses lateral flow immunochro-
matographic assay to detect MTB-specific antigen
123
J.R. Sun, et al
MPB64.33,34 This low-tech rapid test with high sen-
sitivity and specificity can provide a rapid method
for the differentiation of M. tuberculosis from other
mycobacteria in MGIT broth. Disadvantages of the
CapiliaTB kits are also high costs and less routine
usage in clinical laboratories.
IS6110-based inhouse PCR was first reported
by Eisenach et al19 and has been extensively used
in clinical specimens16,17 and MGIT cultures.15 In
this study, the IS6110-based PCR assay was found
to have a sensitivity of 5 cells or 0.1 pg of DNA in
the detection of MTBC. In addition, we obtained
100% correlation between the IS6110-based PCR
and conventional methods in sensitivity and
specificity in mycobacterial identification.
Based on the results of this study, we believe
that IS6110-based PCR can be applied directly to
identify MTBC or NTM in positive MGIT cultures.
This method is less expensive and 3–5 weeks
quicker than the biochemical tests in the identi-
fication of MTB. We therefore recommend that
this combined protocol be used for routine
diagnosis of mycobacterial infections in clinical
laboratories.
Acknowledgments
This study was supported in part by grants from
the Tri-Service General Hospital (TSGH-C94-83
and TSGH-C95-40 to J.J. Lu and TSGH-C95-92
to J.R. Sun). We thank Dr Chao-Hung Lee for as-
sistance with the manuscript.
References
1. Murray SJ, Barrett A, Magee JG, et al. Optimisation of acid
fast smears for the direct detection of mycobacteria in
clinical samples. J Clin Pathol 2003;56:613–5.
2. Abe C, Hosojima S, Fukasawa Y, et al. Comparison of MB-
Check, BACTEC, and egg-based media for recovery of
mycobacteria. J Clin Microbiol 1992;30:878–81.
3. Ichiyama S, Shimokata K, Takeuchi J. Comparative study of
a biphasic culture system (Roche MB Check system) with
a conventional egg medium for recovery of mycobacteria.
Aichi Mycobacteriosis Research Group. Tuber Lung Dis
1993;74:338–41.
124
4. Isenberg HD, D’Amato RF, Heifets L, et al. Collaborative
feasibility study of a biphasic system (Roche Septi-Chek
AFB) for rapid detection and isolation of mycobacteria.
J Clin Microbiol 1991;29:1719–22.
5. Abe C, Hosojima S. Comparison of MB-Check, BACTEC,
and egg-based media for recovery of mycobacteria.
Kekkaku 1992;67:781–6.
6. Tortoli E, Cichero P, Piersimoni C, et al. Use of BACTEC
MGIT 960 for recovery of mycobacteria from clinical
specimens: multicenter study. J Clin Microbiol 1999;37:
3578–82.
7. Idigoras P, Beristain X, Iturzaeta A, et al. Comparison of
the automated nonradiometric Bactec MGIT 960 system
with Lowenstein-Jensen, Coletsos, and Middlebrook 7H11
solid media for recovery of mycobacteria. Eur J Clin Microbiol
Infect Dis 2000;19:350–4.
8. Somoskovi A, Magyar P. Comparison of the mycobacteria
growth indicator tube with MB redox, Lowenstein-Jensen,
and Middlebrook 7H11 media for recovery of mycobacte-
ria in clinical specimens. J Clin Microbiol 1999;37:1366–9.
9. Somoskovi A, Clobridge A, Larsen SC, et al. Does the MGIT
960 system improve the turnaround times for growth de-
tection and susceptibility testing of the Mycobacterium
tuberculosis complex? J Clin Microbiol 2006;44:2314–5.
10. Somoskovi A, Kodmon C, Lantos A, et al. Comparison of
recoveries of Mycobacterium tuberculosis using the auto-
mated BACTEC MGIT 960 system, the BACTEC 460 TB
system, and Lowenstein-Jensen medium. J Clin Microbiol
2000;38:2395–7.
11. Kontos F, Nicolaou S, Kostopoulos C, et al. Multicenter
evaluation of the fully automated Bactec MGIT 960 sys-
tem for susceptibility testing of Mycobacterium tuberculosis
to pyrazinamide: comparison with the radiometric Bactec
460TB system. J Microbiol Methods 2003;55:331–3.
12. Abe C, Hirano K, Wada M, et al. Detection of Mycobac-
terium tuberculosis in clinical specimens by polymerase
chain reaction and Gen-Probe Amplified Mycobacterium
Tuberculosis Direct Test. J Clin Microbiol 1993;31:
3270–4.
13. Stauffer F, Mutschlechner R, Hasenberger P, et al. Detec-
tion of Mycobacterium tuberculosis complex in clinical
specimens by a commercial polymerase chain reaction kit.
Eur J Clin Microbiol Infect Dis 1995;14:1046–51.
14. Ruiz P, Gutierrez J, Zerolo FJ, et al. GenoType mycobacterium
assay for identification of mycobacterial species isolated
from human clinical samples by using liquid medium.
J Clin Microbiol 2002;40:3076–8.
15. Sion C, Degraux J, Delmee M. Early identification of
Mycobacterium tuberculosis and Mycobacterium avium
using the polymerase chain reaction on samples positive
by a rapid commercial culture system. Eur J Clin Microbiol
Infect Dis 1999;18:346–51.
16. Rimek D, Tyagi S, Kappe R. Performance of an IS6110-
based PCR assay and the COBAS AMPLICOR MTB PCR
system for detection of Mycobacterium tuberculosis
J Formos Med Assoc | 2009 • Vol 108 • No 2

complex DNA in human lymph node samples. J Clin
Microbiol 2002;40:3089–92.
17. Caws M, Wilson SM, Clough C, et al. Role of IS6110-
targeted PCR, culture, biochemical, clinical, and immuno-
logical criteria for diagnosis of tuberculous meningitis.
J Clin Microbiol 2000;38:3150–5.
18. Patel S, Wall S, Saunders NA. Heminested inverse PCR for
IS6110 fingerprinting of Mycobacterium tuberculosis
strains. J Clin Microbiol 1996;34:1686–90.
19. Eisenach KD, Cave MD, Bates JH, et al. Polymerase chain
reaction amplification of a repetitive DNA sequence spe-
cific for Mycobacterium tuberculosis. J Infect Dis 1990;
161:977–81.
20. Gillespie SH, McHugh TD, Newport LE. Specificity of
IS6110-based amplification assays for Mycobacterium
tuberculosis complex. J Clin Microbiol 1997;35:799–801.
21. Springer B, Stockman L, Teschner K, et al. Two-laboratory
collaborative study on identification of mycobacteria:
molecular versus phenotypic methods. J Clin Microbiol
1996;34:296–303.
22. Hall L, Doerr KA, Wohlfiel SL, et al. Evaluation of the
MicroSeq system for identification of mycobacteria by 16S
ribosomal DNA sequencing and its integration into a rou-
tine clinical mycobacteriology laboratory. J Clin Microbiol
2003;41:1447–53.
23. Reischl U, Pulz M, Ehret W, et al. PCR-based detection of
mycobacteria in sputum samples using a simple and reli-
able DNA extraction protocol. Biotechniques 1994;17:844–5.
24. Alp A, Hascelik G. Comparison of BBL mycobacteria growth
indicator tube (MGIT) method, BACTEC radiometric system
and Lowenstein-Jensen culture media for the detection of
mycobacteria in clinical specimens. Mikrobiyol Bul 2002;
36:229–35. [In Turkish]
25. Styrt BA, Shinnick TM, Ridderhof JC, et al. Turnaround
times for mycobacterial cultures. J Clin Microbiol 1997;
35:1041–2.
26. Hanna BA, Ebrahimzadeh A, Elliott LB, et al. Multicenter
evaluation of the BACTEC MGIT 960 system for recovery
of mycobacteria. J Clin Microbiol 1999;37:748–52.
J Formos Med Assoc | 2009 • Vol 108 • No 2
View publication stats
Combination of MGIT and PCR for TB diagnosis
27. Pfyffer GE, Welscher HM, Kissling P, et al. Comparison of
the Mycobacteria Growth Indicator Tube (MGIT) with ra-
diometric and solid culture for recovery of acid-fast bacilli.
J Clin Microbiol 1997;35:364–8.
28. Williams-Bouyer N, Yorke R, Lee HI, et al. Comparison of
the BACTEC MGIT 960 and ESP culture system II for growth
and detection of mycobacteria. J Clin Microbiol 2000;38:
4167–70.
29. Scarparo C, Piccoli P, Rigon A, et al. Evaluation of the
BACTEC MGIT 960 in comparison with BACTEC 460 TB
for detection and recovery of mycobacteria from clinical
specimens. Diagn Microbiol Infect Dis 2002;44:157–61.
30. Hillemann D, Richter E, Rusch-Gerdes S. Use of the BACTEC
Mycobacteria Growth Indicator Tube 960 automated sys-
tem for recovery of Mycobacteria from 9,558 extrapul-
monary specimens, including urine samples. J Clin Microbiol
2006;44:4014–7.
31. Kontos F, Petinaki E, Nicolaou S, et al. Multicenter evalua-
tion of the fully automated BACTEC MGIT 960 system and
three molecular methods for the isolation and the identifi-
cation of mycobacteria from clinical specimens. Diagn
Microbiol Infect Dis 2003;46:299–301.
32. Alcaide F, Benitez MA, Escriba JM, et al. Evaluation of the
BACTEC MGIT 960 and the MB/BacT systems for recov-
ery of mycobacteria from clinical specimens and for
species identification by DNA AccuProbe. J Clin Microbiol
2000;38:398–401.
33. Hasegawa N, Miura T, Ishii K, et al. New simple and rapid
test for culture confirmation of Mycobacterium tuberculo-
sis complex: a multicenter study. J Clin Microbiol 2002;
40:908–12.
34. Hirano K, Aono A, Takahashi M, et al. Mutations including
IS6110 insertion in the gene encoding the MPB64 protein
of Capilia TB-negative Mycobacterium tuberculosis iso-
lates. J Clin Microbiol 2004;42:390–2.
35. Katila ML, Katila P, Erkinjuntti-Pekkanen R. Accelerated
detection and identification of mycobacteria with MGIT
960 and COBAS AMPLICOR systems. J Clin Microbiol
2000;38:960–4.
125