| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
in Pleural Fluid for the Differential Diagnosis of Pleural Tuberculosis*
* From the Centro Internacional de Entrenamiento e Investigaciones Médicas, Cali, Colombia.
Correspondence to: Nancy Gore Saravia, MSc, PhD, Centro Internacional de Entrenamiento e Investigaciones Médicas- CIDEIM, Apartado Aereo 5390, Av 1N No. 3–03, Cali, Colombia; e-mail: cideim{at}cali.cetcol.net.co
 |
Abstract |
|---|
 TOP Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
levels in pleural
fluid in the diagnosis of pleural TB.
Patients: ADA
activity, IFN- levels, and PCR were evaluated in 140 cases of
pleural effusion, 42 with confirmed pleural TB, 19 with probable
pleural TB, 70 with a nontuberculous etiology, and 9 having an
undetermined etiology.
Results: ADA activity, IFN-
levels, and PCR were 88%, 85.7%, and 73.8% sensitive, respectively,
and 85.7%, 97.1%, and 90% specific, respectively, for pleural TB
that had been confirmed by either culture or pleural biopsy specimens.
The combination of PCR, IFN- measurement, and ADA activity
determination allowed the selective increase of sensitivity and
specificity for probable and confirmed cases compared to individual
methods. Positive and negative predictive values for these individual
or combined methods were maintained over a wide range of prevalence of
pleural TB in the patient population presenting with pleural effusions.
Fever and younger age were associated with tuberculous pleural effusion
(p < 0.0001), while blood in sputum and older age were associated
with malignant etiology (p < 0.008).
Conclusions:
These clinical variables together with the use of ADA activity
determination, PCR, and measurement of IFN- levels provide the basis
for the rapid and efficient diagnosis of pleural TB in different
clinical settings.
Key Words: adenosine deaminase • diagnosis • interferon- • pleural effusion • polymerase chain reaction • positive and negative predictive values • tuberculosis
|
Introduction |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The paucity of bacilli and the nonspecific cytochemical characteristics of pleural fluid in pleural TB mandate more invasive procedures such as pleural biopsy or thoracotomy for differential diagnosis. Although tuberculous pleural effusion may resolve over a period of several months without treatment, a failure to diagnose and treat pleural TB can result in progressive disease and the involvement of other organs in as many as 65% of patients.7 However, treatment based on clinical suspicion rather than on microbiological diagnosis results in overtreatment, delay in accurate diagnosis, and potentially greater morbidity.
Methods such as the measurement of adenosine deaminase (ADA) activity
in pleural fluid, which is due principally to
ADA2 produced by monocytes8
and is
indicative of a local, active, inflammatory response, the
quantification of interferon (IFN)-,9
10
11
12
13
14
and
polymerase chain reaction (PCR) to detect a specific sequence of the
M tuberculosis genome have shown higher sensitivity than
culturing or the direct examination of pleural
fluid.15
16
17
18
Individually, the measurements of ADA
activity and IFN- levels have proven to be sensitive and specific
for pleural TB in populations with a high prevalence of
TB.9
10
12
15
19
20
However, these diagnostic aids have
not being widely utilized for the routine laboratory diagnosis of
pleural TB. The indirect nature of the evaluation of ADA
activity,21
the yet limited information on IFN- in
pleural effusions, and the risk of false-positive results in
populations with a low prevalence of TB have discouraged the routine
use of these simple tests. However, the high biological sensitivity and
specificity of PCR for M tuberculosis22 suggest
that this method, when used in combination with the measurement of ADA
activity or IFN- levels, could improve the efficiency of the
laboratory diagnosis of pleural TB. To our knowledge, these
combinations have not previously been evaluated.
In this study, we have examined the diagnostic efficiency of ADA
activity, IFN- measurement, and PCR in patients with tuberculous
pleural effusions and related the results to clinical signs and
symptoms in a population with a high prevalence of TB. Subsequently, we
determined the positive predictive values (PPVs) and negative
predictive values (NPVs) of individual and combined methods in relation
to the prevalence of pleural TB in patients with pleural effusions.
|
Materials and Methods |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Laboratory Methods
A single specimen of pleural fluid (50 to 100 mL) was submitted
for cytologic examination, Ziehl-Neelsen staining, ADA activity
determination, measurement of IFN- levels, and PCR. Simultaneously,
a pleural biopsy was performed and the specimen was submitted for
histopathologic examination. Pleural fluid samples were concentrated by
centrifugation for 20 min at 10,000g at 4°C. The pellet
was resuspended in 1.5 mL DNAse-free buffer (pH, 8.0). The sample was
divided to perform direct smears (Ziehl-Neelsen staining) and cultures
in Ogawa-Kudoh medium.23
ADA activity was determined in 1 mL pleural fluid using the colorimetric method described by Giusti and Galanti.24 A positive result was defined as a value > 45.5 U/L based on the previous analysis of pleural fluid samples with proven TB and non-TB. A positive control sample and two negative control samples for which the ADA value was known were included in each group of clinical samples analyzed.
DNA Extraction and PCR
Five hundred microliters of concentrated pleural fluid samples
were lysed in tris-ethylenediaminetetraacetic acid buffer (Tris-Cl, 10
mM; ethylenediaminetetraacetic acid, 10 mM; pH, 8.0) containing 200
µg/mL proteinase K, 0.1% octoxynol-9 (Triton X-100), and 1% sodium
dodecyl sulfate, incubated at 56°C overnight. DNA extraction was
performed using the conventional phenol/chloroform
method25
; DNA was precipitated with 2.5 mM sodium acetate
and ethanol at -20°C overnight. The IS6110 target
sequence DNA was amplified using INS2/Pt18 primers (Arend Kolk; KIT
Institute; Amsterdam, The Netherlands). The amplification products were
evaluated by agarose gel and dot-blot methods described by Kolk et
al.22
26
27
To avoid contamination by amplicons in the PCR
reactions, we used uracil triphosphate and uracil-N-glycosilase enzyme
to eliminate products of prior amplifications.28
Enzyme-Linked Immunosorbent Assay for IFN-
The concentration of IFN- in the pleural effusions was
measured by enzyme-linked immunosorbent assay. Briefly, flat-bottomed
96-well microtiter plates (Immunolon 2; Dynatech Laboratories;
Chantilly, VA) were coated with monoclonal mouse antihuman IFN-
(Genzyme; Cambridge, MA) at 5 µg/mL in sodium carbonate buffer (pH,
9.6 ) and were incubated overnight at 4°C. After blocking with sodium
carbonate buffer containing 10% heat-inactivated fetal calf serum for
2 h at 37°C, serial dilution of standard human IFN-
(Boehringer Mannheim; Indianapolis, IN) and samples were added and
incubated at 37°C for 2 h. Polyclonal rabbit antihuman IFN-
(1:4,000) was added to each well, and the plates were incubated at
37°C for 1 h. Subsequently, peroxidase-conjugated donkey
antirabbit IgG (1:5,000) (Jackson ImmunoResearch Laboratories; West
Grove, PA) was added and was incubated for 30 min at 37°C. The
enzymatic reaction was developed with a peroxidase substrate solution
(Kirkegaard & Perry Laboratories; Gaithersburg, MD), and the plates
were read in an enzyme-linked immunosorbent assay reader (Dynatech
Laboratories) at a wavelength of 410 nm. Between each incubation step,
the plates were washed with phosphate-buffered saline solution
containing 10% fetal calf serum and 0.05% polysorbate 20 (Tween 20)
(Sigma; St. Louis, MO). The IFN- concentration of each sample was
calculated by regression analysis from a standard curve using the mean
absorbance (average of duplicate reading). The sensitivity of this
assay was 0.8 U/mL.
Data Analysis
The results of clinical evaluations, diagnostic tests, ADA
activity determinations, measurements of IFN- levels, and PCRs were
analyzed using computer software (SPSS, version 7.5; SPSS; Chicago,
IL). The sensitivity, specificity, PPVs, and NPVs of ADA activity,
IFN- levels, and PCR results were compared to each of the
conventional methods. The results of the pleural fluid cultures and
histopathologic testing were used as the "gold standard."
The sensitivity and specificity of the diagnostic methods were
determined using the result of each of the diagnostic methods for each
patient. Therefore, the unit of analysis was the patient, not the
clinical specimen or test. Comparisons between the diagnostic methods
were performed using a McNemar 
2 test
(p < 0.05). Student’s t test and analysis of variance
were used to compare the mean production of IFN- among groups. PPVs
and NPVs were estimated using Bayesian analysis29
with a
TB prevalence range of 1 to 50% for the patient population with
pleural effusions.
|
Results |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Seventy-seven percent of the patients with pleural TB (47 of 61) presented with fever compared with 38.6% of patients with malignancies (27 of 70). A statistically significant association was found between pleural TB and fever (p < 0.0001). Productive cough with blood was more frequently present in patients with malignancy (38 of 70 patients; 54%; p < 0.008) than patients with pleural TB (19 of 61 patients; 31%).
Sensitivity, Specificity, and Predictive Values of the Individual
Methods
Measurement of ADA activity was the single most sensitive method
(88.1%) for the diagnosis of pleural TB (Table 1
). The specificity of ADA activity was 85.7% for confirmed and probable
pleural TB combined, a value that is inferior to but approximates that
of culturing, pleural biopsy, and PCR. Nevertheless, ADA activity had
the highest NPV (88.2%), which remained high over a range of
prevalence from 0.01 to 0.5 (Fig 1
).
|
|
IFN- values were significantly higher in the pleural fluid of
patients with confirmed TB (71.7 U/mL) than in those with probable
pleural TB (22.6 U/mL; p < 0.001), yet values for probable as well
as confirmed cases were significantly higher than for those with
nontuberculous pleural effusions (0.7 U/mL; p < 0.001). IFN-
levels were also significantly higher in pleural fluid samples with ADA
activity of 
45 U/L (p < 0.001). IFN- levels in pleural fluid
were slightly less sensitive than ADA activity (85.7%) but were more
specific (97.1%; Table 1
), and the presence of 6 U/mL IFN- had
the highest PPV (94.7%) after the "gold standards" of culture and
biopsy.
The sensitivity of PCR was higher than culturing (p = 0.007) but differed in the confirmed (73.8%) and probable (31.6%) pleural TB groups. The sensitivity of PCR in pleural fluid was comparable to that for pleural biopsy, although specificity was 2 to 10% lower (Table 1) . Dot-blot hybridization increased the sensitivity of PCR with some decrease in specificity compared with the detection of the amplification product by gel electrophoresis. PCR was more sensitive in patients with pleural TB confirmed by culture (18 of 20 patients [85.7%]) than in those with pleural TB confirmed by biopsy specimens who had cultures negative for M tuberculosis (13 of 22 patients [59%]).
Four patients with nontuberculous pleural effusions and three with pleural effusions of undetermined etiology were positive for pleural TB by PCR. Four of these putative false-positive results occurred in individuals with positive results of a PPD skin test, and three occurred in individuals who were PPD negative. The overall frequency of PCR positivity was significantly higher among PPD-positive patients (4 of 21 patients) than among PPD-negative patients (3 of 38 patients) who had confirmed nontuberculous etiologies (p = 0.003). No relationship was found between a history of vaccination with bacillus Calmette-Guérin (BCG) and either PPD positivity or PCR positivity.
The culturing of pleural fluid allowed the diagnosis of pleural TB in 20 of 61 patients (33%) with confirmed probable pleural TB, including 10 patients with pleural TB that had not been detected by pleural biopsy specimens. No samples were positive by Ziehl-Neelsen staining despite culture positivity.
Pleural biopsy was performed in 131 of 140 patients evaluated. The overall sensitivity was 64.3% and specificity was 98% (Table 1) . The sensitivity of histologic testing of pleural biopsy specimens for the diagnosis of malignancy was 72.5%, with 29 of 40 patients receiving confirmations of malignancies.
PPD was applied and read in 117 of 140 patients. Sensitivity was higher in the patient group classified as having probable pleural TB (70.6%) than in the group with confirmed pleural TB (50%). In the patient group having a confirmed nontuberculous effusion, 21 of 59 patients (35%) were PPD positive. Fifty percent of PPD-positive patients had a history of BCG vaccination or scarring from the vaccination. Conversely, among those patients with a history or evidence of BCG vaccination, 25 of 49 were PPD positive.
Cytologic examination was diagnostic in 14 of 40 patients with malignancy, yielding a sensitivity of 35% for this etiology. In patients with TB, cytology presented as undetermined inflammation.
Combination of Diagnostic Methods
The combination of two methods, with a positive result by either
two of the methods considered to be indicative of a positive diagnosis
of pleural TB, increased the diagnostic sensitivity over all individual
methods (Fig 2
, 
Table 3
). Both the measurement of IFN- levels and PCR, when combined with
the determination of ADA activity, yielded high sensitivity (89.7% and
90.2%, respectively) for confirmed and probable pleural TB combined,
whereas IFN- levels in combination with ADA activity presented the
highest specificity (83.8%).
|
|
|
levels and PCR as well as biopsy specimen
plus PCR were 100% specific for confirmed and probable cases of
pleural TB (Table 3)
. The PPVs and NPVs of IFN-/PCR and ADA/PCR were
the highest of all combined methods and were sustained over a wide
prevalence range for TB (Fig 2
, 3
).
|
|
Discussion |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
ADA activity is found in pathologic conditions other than TB.16 30 Nevertheless, high levels of the enzyme are more strongly associated with pleural TB.31 32 Hence, the determination of the cutoff value for the case mix that is included in the differential diagnosis can influence the precision of this diagnostic aid. The cutoff value utilized in this study (45.5 U/L) is similar to that established in other settings (33 to 47 U/L) where ADA has been evaluated in the differential diagnosis of pleural TB.33 34 35 In contrast with the indirect nature of ADA activity, PCR, which is based on the amplification of a specific genomic sequence of M tuberculosis, is theoretically highly specific. Contamination either with extraneous DNA or with amplification products of previous analyses is the principal source of false-positive results. In this study, the use of uracil triphosphate and uracil-N-glycosylase enzyme to eliminate amplicons28 in addition to negative control samples and procedures to eliminate cross-contamination, together with the high biological specificity of PCR,26 support the interpretation that amplification products in samples from patients with probable TB, other etiology, and undetermined etiology indicate the presence of M tuberculosis. The higher frequency of PPD positivity, which is indicative of prior infective exposure to M tuberculosis, among patients with PCR positivity, compared to those with PCR negativity, is consistent with the possibility that these patients experienced either latent infection and pulmonary or pleural granulomas without clinical manifestations or the reactivation of infections that are coexistent with the immunosuppressive conditions causing pleural effusions, such as cardiac or renal insufficiency and metastatic malignancy.
The frequency of false-negative results of PCR in this study (26%) is most likely attributable to the inefficient recovery of genomic DNA from the characteristically low number of bacilli in patients with pleural TB. The extraction of DNA from frozen samples, as occurred in this study, can be less efficient than extraction from fresh samples,36 and the optimization of the extraction step improves the sensitivity of the application of PCR to paucibacillary presentations of TB.37 Since inhibition was routinely monitored in the samples, inhibition is unlikely to have significantly diminished the sensitivity of the PCR assay. However, some strains of M tuberculosis in Asia lack the IS6110 sequence.38 39 Although the absence of this sequence in some strains of M tuberculosis could cause false-negative results, such strains are infrequent and have not yet been reported in South America.
The high sensitivity and specificity of ADA in the differential diagnosis of pleural TB have been substantiated in studies conducted in other areas of high TB prevalence, such as Mexico, Spain, and Brazil.15 19 20 Furthermore, using Bayesian analysis, it was possible to demonstrate that ADA sustains high PPVs and NPVs over a relatively wide spectrum of prevalence of this presentation of TB. This theoretical projection encourages the evaluation of ADA as a diagnostic aid for the differential diagnosis of pleural TB in populations with moderate to low prevalence of TB. The reported increase of ADA activity in patients with parapneumonic effusions will be an important consideration in evaluating the diagnostic utility of ADA activity in populations having a low prevalence of TB.16 Nevertheless, the sensitivity and robustness of ADA activity in the diagnosis of pleural TB, together with its simplicity, speed, and low cost, urge the widespread implementation and routine utilization of the method in populations with a high prevalence of TB and its evaluation in areas of low prevalence.
IFN-, a cytokine associated with a Th1 type of cell-mediated immune
response, proved to be highly associated with a tuberculous etiology of
pleural effusion. The clear separation of 95% confidence intervals for
values of this cytokine in the pleural fluid of patients with confirmed
and probable pleural TB from those with nontuberculous pleural
effusions, and the high sensitivity, specificity, and predictive values
observed for IFN- levels in this case series substantiate its
diagnostic potential. These findings are consistent with the published
experience on the presence of IFN- in the pleural fluid of patients
with pleural TB compared with patients with other etiologies of pleural
effusion.10
11
12
14
The sensitivity of PCR in pleural fluid testing in this case series
(culture-positive patients, 85.7%; biopsy specimen-diagnosed patients
with negative cultures, 59%; patients with confirmed pleural TB, 73%)
was comparable to that found in other evaluations of this
method for the diagnosis of pleural TB.6
17
22
37
40
41
42
Among probable cases, the 31% that could be confirmed as due to
M tuberculosis by PCR represents a gain over the "gold
standards." The sensitivity gain in etiologic diagnosis and the high
PPVs and NPVs of PCR substantiate the diagnostic utility of this
method. In addition, PCR offers the advantage of speed in obtaining
results and the option of referring the sample rather than the patient
to a specialized center or laboratory. Comparison of the sensitivity
and specificity of PCR with those of culture and biopsy, the "gold
standards," supports the feasibility of using the PCR of pleural
fluid together with determination of ADA activity and measurement of
IFN- levels as a first diagnostic approach in circumstances in which
pleural TB is suspected and access to procedures such as needle biopsy
of the pleura is limited. This would facilitate the opportune
identification of patients who would benefit from treatment. However,
when access to pleural biopsy is not an issue and prevalence of TB is
low, determination of ADA activity, measurement of IFN- levels, and
PCR of pleural fluid are valuable adjunct procedures. In either
circumstance, earlier diagnosis and treatment are favored by the
implementation of ADA, IFN-, and PCR analyses of pleural fluid.
Although PCR, IFN- measurement, and ADA activity have been evaluated
individually in the diagnosis of pleural TB, the results of using all
three methods in the same patients have not been available previously,
and their combined use has not been explored. Our findings show that
together, measurement of IFN- levels, PCR, and determination of ADA
activity can be utilized to preferentially optimize sensitivity, in an
either/or combination, or specificity, in a combination requiring both
methods used to be positive. Therefore, the challenge of
diagnostic efficiency in different circumstances of prevalence may be
addressed using combinations of these rapid methods on pleural
fluid.
Cost and expertise are also considerations in adopting technologies for
use at different levels of health service. ADA and IFN- measurement
are simple and comparatively low-cost procedures, whereas PCR is a more
demanding and expensive method yet has the advantage of being a rapid
and direct means of detecting M tuberculosis in pleural
fluid. We believe that the results of this study demonstrate that
individually and in combination, these methods can offer a
cost-effective means of obtaining diagnostic efficiency.
The significant associations observed in this study between fever and tuberculous pleural effusion and hemoptysis or productive cough with a malignant etiology could be exploited in the development of a clinical prediction rule that also included laboratory findings for ADA activity and PCR to discriminate between these two frequent causes of pleural effusion. Age was also associated with the different etiologies; tuberculous effusion was associated with younger age, while malignancy was associated with older age. This variable also could be included in such a rule. Young age has been associated previously with the diagnosis of tuberculous pleural effusion; in the study by Valdes and collaborators,20 it was considered together with the infrequency of pulmonary TB in the affected age group to suggest that in cases of pleural effusion in younger persons, pleural TB may be a primary form of TB. Furthermore, the prevalence of tuberculous effusion reportedly decreases with age while no relation was observed between ADA activity in pleural fluid and age.43
In the present investigation, conventional and alternative diagnostic methods were evaluated individually and in combination within the context of the case mix encountered in clinical practice. We have explored combinations of methods the complementary strengths of which offer diagnostic options that can be utilized advantageously in different technologic, epidemiologic, and clinical circumstances. The understanding of the strengths and weaknesses of each diagnostic method, used individually or in combination, and the consideration of the results together with the symptoms and signs associated with a determined etiology of pleural effusion propitiates the rational and optimal election of methods in different diagnostic scenarios. The results of this study contribute a comprehensive view of the comportment of the repertoire of methods currently in use or available but not routinely utilized in the differential diagnosis of pleural effusion. These findings support the use of new and more efficient diagnostic strategies in the management of pleural TB.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
This research was supported by COLCIENCIAS grant No. 2229–04-184–95.
Received for publication October 22, 1999. Accepted for publication April 11, 2000.
|
References |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, ADA and CA 125 as the diagnostic parameters in tuberculous pleuritis. Respir Med 88,139-143[CrossRef][ISI][Medline]
This article has been cited by other articles:
|
|
 |
|
  A. Gopi, S. M. Madhavan, S. K. Sharma, and S. A. Sahn Diagnosis and Treatment of Tuberculous Pleural Effusion in 2006 Chest, March 1, 2007; 131(3): 880 - 889. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. S. Wiener, P. Della-Latta, and N. W. Schluger Effect of Nucleic Acid Amplification for Mycobacterium tuberculosis on Clinical Decision Making in Suspected Extrapulmonary Tuberculosis Chest, July 1, 2005; 128(1): 102 - 107. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. A. Gupta, S. K. Chhabra, A. Hiraki, and K. Aoe Diagnosing Tubercular Pleural Effusions Chest, March 1, 2005; 127(3): 1078 - 1079. [Full Text] [PDF]
|
|
|
|
|
|
N. A. Hasaneen, M. E. Zaki, H. M. Shalaby, and A. S. El-Morsi Polymerase Chain Reaction of Pleural Biopsy Is a Rapid and Sensitive Method for the Diagnosis of Tuberculous Pleural Effusion Chest, December 1, 2003; 124(6): 2105 - 2111. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. M. Lima, J. K. B. Colares, and B. A. L. da Fonseca Combined Use of the Polymerase Chain Reaction and Detection of Adenosine Deaminase Activity on Pleural Fluid Improves the Rate of Diagnosis of Pleural Tuberculosis Chest, September 1, 2003; 124(3): 909 - 914. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Aoe, A. Hiraki, T. Murakami, R. Eda, T. Maeda, K. Sugi, and H. Takeyama Diagnostic Significance of Interferon-{gamma} in Tuberculous Pleural Effusions Chest, March 1, 2003; 123(3): 740 - 744. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Romano, M. Sanguinetti, B. Posteraro, F. Ardito, G. Gesu, A. M. Schito, and G. Fadda Early Detection of Negative BACTEC MGIT 960 Cultures by PCR-Reverse Cross-Blot Hybridization Assay J. Clin. Microbiol., September 1, 2002; 40(9): 3499 - 3501. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
