Tuberculosis (TB) is the number one infectious disease killer worldwide.
The World Health Organization estimates that 2 billion people have latent
TB, while another 3 million people worldwide die each year due to TB (Thomas
Herchline, 2007). TB can affect almost every organ in a human`s body but
pulmonary TB is the most common. The high risk of person-to-person transmission,
morbidity and mortality makes it one of the major threats to public health
worldwide. Early diagnosis followed by adequate treatment is essential
to prevent the threat. Tuberculous pleural effusion occurs in 30-60% patients
throughout the world (Angeline et al., 2007). The conventional
laboratory method for detection of Mycobacterium requires weeks for completion.
Molecular method promises rapid detection and identification of Mycobacterium
directly from clinical samples. Currently, there is no clear guideline
for usage of molecular assay in diagnosis of extra pulmonary TB. Therefore
the present study is aimed at to determine the sensitivity and specificity
of PCR in detecting M. tuberculosis from pleural fluid and
comparison with other diagnostic methods.
MATERIALS AND METHODS
Study design:The study was performed prospectively in a blinded manner, where
the final clinical diagnosis was not available at the time when the specimens
Study population:All pleural fluid sample sent to Microbiology Laboratory, Hospital
University Kebangsaan Malaysia for routine culture between May 2005 to
October 2006 were eligible for enrollment in this study. Samples collected
from other sites such as sputum and bronchi alveolar lavage were also
included in this study. Subsequently, the patients were placed under tuberculous
(TB) and non-tuberculous (Non-TB) group based on the data collected from
patient case files. All patients in the TB group was treated with anti-tuberculous
treatment and showed marked improvement.
Sample size:The sample size for this study was calculated by using the Z-test
formula. Where Z equals to 1.96 (80% power of study), expected population
proportion (prevalence) of 0.04, absolute precision of 0.10 and at 95%
confidence interval, the minimum sample size expected 15.
Sample collection and processing:All pleural fluid received was divided into two parts. One part
was centrifuged and processed following the standard operating procedure
of laboratory for smear preparation on which Kinyoun stain was done and
also cultured on Löwenstein-Jensen (LJ) medium. The inoculated LJ
medium was incubated at 37°C for 42 days. The other half was used
for Nested PCR.
Clinical data collection:Patient profiles and clinical data were retrieved from the case
file of the medical record office. These data were manually entered in
a form. The data were finally entered on a master sheet.
Nested PCR assay
Primers:The ABSOLUTETM MTB II PCR kit by BioSewoom (Seoul, Korea)
amplifies the IS6110 fragment of M. tuberculosis, a conserved
repetitive sequence in the M. tuberculosis complex DNA.
Pretreatment of specimens:An amount of 0.5 mL of pleural fluid was centrifuged at 12,000 rpm
for 2 min. The supernatant was discarded. Eight hundred microliter of
distilled water was added to the pellet. The specimen was again centrifuged
at 12,000 rpm for 2 min and the supernatant discarded. Finally, 800 μL
of distilled water was added to the remaining pellet. The specimen was
again centrifuged at 12,000 rpm for 3 min and the supernatant discarded.
DNA extraction:The pellet from the pretreated sample was mixed with 50 μL
of DNA extraction buffer. The mixture was placed in a heat block at 56°C
for 15 min, then vortexed at high speed for 10 sec. Next, the mixture
was placed in a heat block at 100°C for 8 min. Then again vortexed
at high speed for 10 sec and followed by centrifugation at 12,000 rpm
for 3 min and the supernatant was used for the amplification step.
Amplification of M. tuberculosis DNA:Nested PCR kit was used for amplification of DNA. The first PCR
master mixture was prepared by adding 15.0 μL of first PCR mixture
with 0.5 μL of first PCR enzyme. A 4.5 μL DNA extract was added
to the first PCR master mixture and mixed well. The samples were subjected
to first PCR condition as described in Table 1.
The second PCR master mixture was prepared by adding 18.0 μL of
second PCR mixture with 0.5 μL of second PCR enzyme. This master
mixture was then added with 1.5 μL of the first PCR product and mixed
well. The samples were subjected to second PCR condition as described
in Table 2.
||Step of amplification for first PCR condition
|| Step of amplification for second PCR condition
Positive and negative controls were included in each run of PCR. Distilled
water was used to replace the DNA extract as a negative control. Positive
control was provided with the kit.
Detection of nested PCR product:A 5 μL portion of second PCR product was added to 1 μL
of loading dye and subsequently was electrophoresed on a 2% agarose gel
(Promega Corporation, USA, catalogue number V3121) in Ultra PureTM
TBE buffer (GIBCO BRL, USA, catalogue number 15546-013) at 100 volt for
30 min. The gel was stained with ethidium bromide and the 183 bp specific
amplified bands were visualized on an ultraviolet transilluminator. GeneRulerTM
100 bp Ladder Plus (MBI Fermentas, USA, catalogue number SM0328) was used
to identify the specified amplified fragment.
RESULTS AND DISCUSSION
Patients`s profile:During 18 months study period, 67 pleural fluid samples were included
in the study. Based on data collected from patient`s clinical record 21
patients were grouped into final diagnosis of TB and 46 were diagnosed
other than TB. The latter group was regarded as control group (Table
In the tuberculous group there were 7 females and 14 males. They were
placed into this group after final diagnosis of pulmonary TB on the basis
of histological study, staining and cultural characteristics of the organisms,
radiological findings and/or clinical improvement (Table
3). Age ranges of the patients were between from 17 to 84 years with
a mean age of 48.8 years (Fig. 1). Ten Chinese, 8 Malays
and 3 non-Malaysian were recorded in this group (Fig. 2).
The ages of the patients in non- tuberculous group ranged between 14
and 87 years with a mean age of 54.4 years. Eighteen females and 28 males
were included in this group (Fig. 1). Four Indians, 20
Malays, 21 Chinese and 1 non-Malaysian were in this group (Fig.
||Profile of the patients included in the study
|*:Group with suggestive chest radiographic findings
and marked improvement after antituberculosis treatment
#:Control group: Systemic lupus erythematosus, congestive cardiac failure, liver disease,
lymphoma, carcinoma and other
||Distribution of patients by age
||Distribution of patients by race
||Results of acid fast staining, culture and nested PCR
|*Note: One sample for culture was contaminated
Acid-fast staining of pleural fluid:
All pleural fluid samples in the TB group were found to be AFB negative.
None of the samples from the non-TB group was AFB positive (Table
). Those samples taken from other sites, 2 (sputum samples) were
AFB positive. Thus, AFB staining method was not able to detect any bacteria
from pleural fluid even in case of 100% positive samples. Thus the sensitivity
of AFB stain on pleural fluid was 0% with a specificity of 100%. The detection
rate of M
by AFB staining was 0%.
Culture of pleural fluid:Only one pleural fluid produced typical colonies in the specialized
media M. tuberculosis. Another sample was found to be contaminated.
All samples in the non-tuberculous group were cultured negative (Table
4). The detection rate of M. tuberculosis in pleural
fluid by culture was 1.5%. The sensitivity of culture in this study showed
4% with a specificity of 100%. The positive predictive value was 100%
with 71% negative predictive value.
Nine patients exhibited positive cultured results from sputum samples,
including the patient with the positive culture from pleural fluid. Thus,
pleural fluid culture could not detect 89% of positive samples.
Figure 3 showed gel documentation of electrophoresis
for the detection of the nested PCR product of TB. The size of M.
tuberculosis specific amplified product is 183 bp. Four samples
showed positive band at the 183 bp position. Distilled water was used
as the negative control. Positive control was observed as a broad band
at position 183 bp.
Using IS6110 insertion sequence, M. tuberculosis complex
DNA was detected in 6 patients. Four of them were in the tuberculous group
and the remaining 2 were in the non-tuberculous group (Table
4). The detection rate of M. tuberculosis in pleural
fluid by TB PCR was 9%. The sensitivity of PCR was 19% with 96% specificity.
PCR results exhibited 67% positive predictive value and 72% negative predictive
None of the pleural fluid samples in the tuberculous group were both
PCR and culture positive. Five of the PCR positive samples were cultured
negative and 1 was contaminated. Nine patients in the tuberculous group
grew M. tuberculosis in samples taken from other sites,
i.e., sputum. Among the 4 true positive PCR, one patient`s
||Detection of PCR product of Mycobacterium tuberculosis.
M: 100 bp Marker, 1: Patient 45, 2: Patient 46, 3: Patient 47, 4:
Patient 48, 5: Patient 49, 6: Patient 50, 7: Patient 51, 8: Positive
control, 9: Negative water
sputum was AFB positive but cultured as negative. Another patient had
sputum AFB negative but culture gave positive results. The remaining two
patients were sputum AFB and culture proved negative. They were both clinically
diagnosed as TB. One patient had family history of TB and the other had
radiological evidence of pulmonary TB. Both patients respond to therapeutic
The high number of false negative result in this study is a matter of
great concern. There are few possible reasons for false negative result
in pleural effusion samples. The pleural effusion is a result of possible
hypersensitivity reaction as mentioned earlier (Nagesh et al.,
2001). The paucibacillary nature of the disease, the availability of small
amount or volume of sample and non-uniform distribution of microorganism
during aliquoting of samples further contributes to the number of false
False negative result causes reduction in sensitivity of any test. A
false negative result in any molecular technique may be due inhibitor
present in sample. Improving the extraction technique and the use of internal
control to determine efficiency may prevent this. Another reason is the
specific infecting organism that was detected by the primer was not probably
present in the samples. This problem may be overcomed by increasing the
sample number from each patient to cover more types and time points (Huggett
et al., 2003). The absence of IS6110 might have also caused
a false positive result. The occurrence of M. tuberculosis
strain deficient of IS6110 has been reported worldwide such as 1% in San
Francisco, Vietnam (2%) and Chennai (4%) (Radhakrishnan et al.,
2001).The author mentioned that 62.5% of the study population in Kerala
was not typeable by IS6110- based fingerprinting.
The diagnosis of extra pulmonary tuberculosis is still challenging for
a number of reasons. The lack of adequate sample volumes, non- uniform
distribution of microorganism contributes to this problem. The paucibiliary
nature of the specimen and the presence of inhibitors undermine the performance
of molecular techniques (Pfyffer et al., 1996).
The result of this study shows a detection rate of 9% by PCR compared
to 0% by AFB staining and 1.5% by culture. This suggests that PCR is a
more sensitive method for detection of TB from pleural fluid (19%) compared
to AFB stain (0%) and culture (4%). However, the test singly is not enough
to identify all cases of TB. However, it can be used as an adjunct to
the traditional methods of laboratory diagnosis of TB.
In conclusion, TB PCR is a rapid method in the diagnosis of tuberculous
pleural effusion. Molecular detection of M. tuberculosis
is very useful in cases that are highly suspected as pleural tuberculosis
those are negative for AFB and culture. However, because of the relatively
low sensitivity of TB PCR in pleural effusion, clinical judgment remains
the ultimate decision in the management of tuberculosis.
Authors acknowledge the help of Mr. Rosmadi and Miss Aini Hayati,
Department of Microbiology and Immunology, Faculty of Medicine, UKM during
molecular works of the present study.