ABSTRACT

1
U-057
TK SYSTEM The Colorimetric Mycobacterial Culture
System Enables Rapid, Easy and Effective
Diagnosis of Tuberculosis
Tanil Kocagöz1, Sema Tiryaki2, Thomas Silier3,
Cengiz Güney4
1Yeditepe University Med. Sch., Istanbul, TURKEY,
2Turkish Innovative Biotechnology Organization,
Istanbul, TURKEY, 3SALUBRIS, Inc., Woburn, MA,
U.S.A., 4Ministry of Health, Sureyyapasa
Pulmonary and Cardiovascular Diseases Education
and Research Hospital, Istanbul, TURKEY.

• ABSTRACT
• Isolation of mycobacteria from clinical samples
  is the most sensitive and definite way of
  tuberculosis diagnosis. When classical
  mycobacterial culture media like Löwenstein
  Jensen (LJ) are used, it is required 3 to 6 weeks
  for the detection of growth. All previously
  developed rapid mycobacterial culture systems use
  selective Middlebrook broth and suffer from not
  being ready-to-use, since this medium requires
  the addition of oleic acid, albumin, dextrose,
  catalase (OADC) and selective antimicrobials,
  before inoculation. They also can not
  differentiate mycobacterial growth from
  contamination prior to microscopic examination
  since it is not possible to understand what
  causes turbidity when there is growth in a liquid
  medium. We have previously developed TK
  (Salubris, Inc.), a ready-to-use rapid culture
  medium which indicates mycobacterial growth by
  changing its original red color to yellow and
  growth of contaminating organisms by a color
  change to green. We have tested the efficiency of
  TK and its selective type TK SLC, which contains
  selective antimicrobials, in the isolation of
  mycobacteria from clinical samples and compared
  it to Löwenstein Jensen and BACTEC MGIT 960
  (Becton Dickinson). After decontamination and
  concentration of 261 samples by NaOH-NALC method
  using Mycoprosafe (Salubris, Inc.) we have
  inoculated them to LJ, MGIT, TK and TK SLC. While
  the growth in LJ was followed visually, MGIT
  tubes were followed in MGIT 960 instrument, TK
  and TK SLC in automated incubator reader MYCOLOR
  TK. Mycobacterial growth was detected in 71
  samples (27) in at least one of the media
  inoculated. Contamination was observed in 1.53
  of LJ, 1.15 of MGIT, 2.68 of TK and 0.77 of TK
  SLC. The median for growth detection time was 23
  days for LJ, 10 days for MGIT, 14 days for TK and
  TK SLC. Having the advantages of being ready to
  use, the ability of differentiating mycobacterial
  growth from contamination and providing growth
  results in average 40 earlier than LJ medium, TK
  culture system proves to be a new effective
  culture system especially for laboratories
  processing large number of samples.

INTRODUCTION Tuberculosis continues to be one of
the major health problems around the world. The
most important element in tuberculosis control is
to find out patients who have tuberculosis
bacilli in their sputum and treat them before
they transmit the disease to healthy individuals.
Culture is the gold standard, most sensitive and
specific method for the diagnosis of
tuberculosis. Unfortunately it takes 3 to 6 weeks
to detect mycobacterial growth in classical media
like Löwenstein Jensen (LJ). Rapid automated
culture systems like BACTEC MGIT 960 (BD
Diagnostic Systems, USA) and BacT/Alert 3D
(Biomerieux, France) are not used widely as
classical media, mainly because of their high
cost, requirement for expensive instrumentation
and their media being not ready to use. All rapid
mycobacterial culture systems except TK use
Middlebrook broth that require the addition of
oleic acid, albumin, dextrose, catalase (OADC)
and selective antimicrobials before inoculation
of processed sample. TK Medium is a ready to use,
rapid mycobacterial culture medium. It enables
early detection of mycobacterial growth by
changing its color. Its original red color turns
to yellow by mycobacterial growth. The color
change occurs before the colonies become visible.
TK Medium has the advantage of differentiating
mycobacterial growth from the growth of most
common contaminants like fungi and Gram negative
bacilli by changing to green instead of yellow
when these microorganisms grow (Figure 1). It is
designed as a solid medium to enable the
visualization and isolation of individual
colonies. Pure cultures of mycobacteria can be
obtained by subculturing individual mycobacterial
colonies if mixed organisms are obtained in the
original culture. Since the growth in TK Media is
detected by color change, it can be easily
followed by visual evaluation and can be used in
laboratories that have a regular 37C incubator.
On the other hand it has a very eloborate and
inexpensive automated incubator reader, called
Mycolor TK (Salubris Technica) (Figure 2 and 3).
Mycolor TK, provides growth curves and its expert
system predicts the type of growing
microorganism, without the need for expensive
additional software. Mycolor TK also enables the
recording of all detailed information related to
the sample and patient including the result of
microscopy. It provides easily statistical data
like the rate of culture positives in different
samples, resistance rate for each
antimycobacterial drug and multi-drug resistant
(MDR) isolates. This study was done to
investigate the performance and advantages of TK
culture system in daily use in a busy
tuberculosis diagnostic laboratory.
RESULTS Among 261 samples, 40 were found to be
AFB positive by microscopic examination.
Mycobacterial growth was detected in 71 samples
(27), in at least one of the media inoculated.
Some isolates grew only in some type of media and
did not grow in others. The microscopy results
of culture positive samples are shown in table 1.
The growth of mycobacterial isolates according to
the type of media is shown in table 2. The median
for growth detection time was 23 days for LJ, 10
days for MGIT, 14 days for TK and TK SLC.
Contamination was observed in 1.5 of LJ, 1.2 of
MGIT, 2.7 of TK and 0.8 of TK SLC. The results
showing the performance of each medium are shown
in table 3.
Table 3. The performance of culture media in the
diagnosis of tuberculosis. Culture Medium
Mycobacteria Growth detection
Contamination
isolation time (median, days)
MGIT 25.3
10
1.2 TK
23.0
14 2.7 TK
SLC 23.8
14 0.8
LJ 23.8
23
1.5
Figure 1. TK Medium turns yellow by
mycobacterial growth and green by
contamination
DISCUSSION Culture is the gold standard method in
the diagnosis of tuberculosis. Although it takes
several weeks to detect mycobacterial growth in a
classical culture medium, its sensitivity and
specificity is much better than microscopy. Rapid
mycobacterial culture systems developed so far
did not manage to replace classical culture media
like LJ, because of several disadvantages among
which their high cost, requirement of additional
instrumentation, and most important, being not
ready to use can be included. To overcome the
disadvantages present in other rapid culture
systems, we have previously developed the TK
culture system. This study was done to evaluate
the performance of TK culture system in the
diagnosis of tuberculosis. Our study site was a
busy tuberculosis diagnostic laboratory with high
percentage of AFB positive samples. However the
concentration of AFB in the samples was pretty
low. Among all 71 culture positive samples, 63
were either smear negative (31) or AFB 1 (32).
The growth detection time depends on the
concentration of mycobacteria in the sample,
especially in rapid culture systems that detect
the metabolic activity. Median growth detection
time, obtained in this study, can be considered
very good for MGIT (10 days) and TK Media (14
days) since the concentration of mycobacteria in
the samples were pretty low. The contamination
rate was low in all types of media including TK
Medium which does not contain selective
antimicrobials. This can be considered due to
effective decontamination and concentration by
Mycoprosafe. This may have played also an
important role in speeding up the growth
detection even in LJ which was 23 days in average
in this study. Although TK Media were slightly
slower than MGIT, they provided culture results
40 earlier than LJ medium. On the other hand it
was much easier to use TK and LJ media since they
were ready to use and did not require preparatory
work like MGIT before inoculation of samples. TK
culture system was also the easiest to follow
since the system can differentiate real
mycobacterial growth from contamination and all
tubes showing growth or negative tubes that
completed incubation duration are indicated on
the screen of Mycolor TK and it takes only a few
minutes every day to evaluate the results.
Having the advantages of being ready to use, the
ability of differentiating mycobacterial growth
from contamination, providing growth results
earlier than classical culture media, and being
very practical to follow the culture tubes, TK
culture system proves to be a new effective
culture system in the diagnosis of tuberculosis.
Table 1. Microscopy results of culture positive
samples. AFB
Negative 1 2 3
4 Number of culture positive samples (71)
31 32 2
5 1
TK MEDIUM MYCOBACTERIA
CONTAMINATION
Table 2. Mycobacterial isolates obtained
in different culture media. Types of
Media in which
Number of mycobacterial isolates are obtained
mycobacterial isolates Growth on
LJ, MGIT, TK and TK SLC
59 MGIT, TK and TK SLC only
1 MGIT
and LJ only
2 TK and TK SLC
only
1 LJ and TK SLC only

1 MGIT only

4 LJ only

1 TK SLC only

2 Total
71
MATERIALS AND METHODS Samples submitted to the
microbiology laboratory of Süreyyapasa Pulmonary
and Cardiovascular Diseases Education and
Research Hospital, Istanbul, Turkey were included
in the study. A total of 261 sputum samples were
processed by NaOH-NALC decontamination and
concentration method, using the ready to use kit
Mycoprosafe (Salubris Inc., USA). MGIT tubes were
prepared by the addition of OADC and PANTA as
suggested by the producer. From processed samples
0.5ml was inoculated to MGIT and LJ tubes and
0.2ml to TK Medium and TK SLC. MGIT tubes were
incubated in BACTEC MGIT 960 instrument. TK Media
were followed by automated incubator reader
Mycolor TK. LJ tubes were incubated in a regular
37C incubator. LJ tubes were checked three times
a week for mycobacterial colonies. As soon as
colonies were visible, this was recorded as
growth detection time for LJ. The growth
detection time for MGIT and TK Media were
recorded as indicated by the automated
instruments. Smears were prepared from any type
of media that indicated growth, stained by Ziehl
Neelsen method and checked by microscopy for the
presence of acid fast bacilli (AFB) and other
contaminating organisms.
Figure 3. Mycolor TK
Acknowledgement This study was supported by
Foundation for Innovative New Diagnostics (FIND).
Figure 2. Mycolor TK
ASM General Meeting. May 21-24, Toronto, Canada