Tuberculous pleurisy is the most common cause of pleural disease in many countries. Even with newer diagnostic tools, it remains a diagnostic challenge and also a subject of interest to researchers. Diagnosis is supported by clinical manifestations, imaging, biological markers, and the identification of TB DNA or acid fast bacilli in pleural fluid. Bacterial culture from pleural fluid and histopathological examination performed from biopsied pleural tissue are considered the gold standard for the diagnosis. The role of pleural fluid biomarkers in the diagnosis of tuberculous pleurisy represents a real interest for the majority of the studies. Studies that followed the accuracy of different biomarkers in tuberculous pleural effusions showed the highest specificity and sensitivity for IFN-γ and total adenosine deaminase (ADA) levels. In clinical practice both biomarkers may be useful to avoid the invasive diagnostic procedures; the significant difference of the test price makes ADA the most preferred diagnostic test. The ADA2 isoenzyme is responsible for the increased activity of ADA in tuberculous pleurisy. Being also markers of the inflammatory process in the pleural space, they can’t establish by itself the tuberculous etiology. Composite scoring systems were developed to facilitate the diagnosis of tuberculous pleurisy and also to avoid expensive, invasive and time-consuming procedures.
Tuberculous pleurisy is the most common cause of pleural disease, and in many countries represents the main form of extrapulmonary tuberculosis(1,2,3). Isolated in 1882 by Robert Koch, Mycobacterium tuberculosis (the Koch bacillus) is the pathogen involved in the develop-ment of the infection and is part of the Mycobacteriaceae family. Even though in many countries the treatment and vaccination programs are available, tuberculosis (TB) remains a major public health problem in develop-ing countries(4). A mycobacterial etiology should be con-sidered in any pleurisy otherwise undiagnosed(2).
The first step in the pathogenesis of primary tubercu-lous pleurisy is the rupture of a lung subpleural caseous granuloma into the pleural space 6-12 weeks after an initial infection. Mycobacterial antigens enter the pleural space
and interact with immune cells previously sensitized to mycobacteria. Thus, the pleural fluid accumulates as a result of a delayed proinflammatory hypersensitivity reac-tion, by increasing the permeability of pleural capillaries to serum proteins. This process may develop in both pri-mary and reactivation of tuberculosis and the pleural space might not always be populated by viable bacilli(1,2,5).
Even if the diagnosis of tuberculous pleurisy remains today a subject of interest to researchers, it is still a diagnostic challenge. Diagnosis is supported by clinical manifestations, imaging, biological markers and iden-tification of TB DNA or acid fast bacilli in pleural fluid. Bacterial culture from pleural fluid and histopathologi-cal examination performed from biopsied pleural tissue are considered the gold standard for the diagnosis.
Tuberculous pleurisy commonly presents in the form of an acute illness, but in a few cases, the onset of symp-toms is slowly progressing, weeks before the diagnosis. Elderly and HIV-positive patients have a particularly long symptomatic or oligosymptomatic period. Unlike patients with pulmonary tuberculosis, patients with tuberculous pleurisy tend to be younger(2). The likeli-hood that a unilateral pleural effusion is of tuberculous etiology decreases with age, because older patients tend to have other comorbidities that could be responsible for the appearance of their pleural effusion, such as pleural malignancy(5).
The most common symptoms are nonproductive cough and pleuritic pain and if both are present, the pain precedes the cough. Additional symptoms are repre-sented by fever, night sweats, weight loss, weakness, fatigability and, last but not least, patients may have dyspnea whose intensity may vary depending on the amount of accumulated pleural fluid. These symptoms are not specific, because they can accompany the clinic of any pleural effusion whose etiology is not tuberculous(3,4,5).
Conventional chest radiography represents the cheapest and most frequently used imaging tool. It is necessary for a minimum of 150 ml of fluid to be accu-mulated in the pleural space so that standard chest X-ray can detect it(6). Usually, it reveals unilateral pleural effusions with a slight right-sided predominance. The amount of pleural effusion varies from small to moder-ate, in the majority of the cases the hemithorax being occupied less than one-third. The opacification of whole hemithorax by the presence of massive amount of liquid would rather favor differential diagnosis like malignancy(3,6).
Chest computed tomography (CT) scan offers addi-tional information and improves the diagnostic accuracy by highlighting associated parenchymal lesions and lim-phadenopaty(2). The prevalence of associated pulmonary parenchymal lesions appears to be higher using CT scan as diagnostic imaging tool as compared to conventional chest X-ray(6).
Pleural ultrasonography is well recognized to be use-ful in performing guided thoracentesis or closed pleural biopsies. The identification of the pleural thickening or other pleural abnormalities may guide the physician to the preferred biopsy site. Also, the ultrasonographic appearance of the effusion gives additional information about its nature(3).
It was previously considered that patients diagnosed with tuberculous pleurisy without the coexisting paren-chymal infiltrates are sputum negative, and thus with-out the risk of transmitting the infection. The quality of the sputum’s samples is of crucial importance. Also, it is very important to obtain cultures of induced sputum samples even in the absence of associated paren-chymal disease because the likelihood to obtain diagnostic confirmation from sputum examination increases using samples obtained by sputum induction. Mycobacterial analyses of sputum have low sensitivity (0-30%) and in the absence of pulmonary lesions the sensitivity decreases even more (4-7%)(2,3).
Pleural fluid characteristics
An important role in the diagnosis of tuberculous pleural effusions is played by the examination of pleural fluid. If sufficient to allow safe puncture, thoracentesis is a mandatory diagnostic step. Usually, the fluid appears clear and yellow colored, except for tuberculous empyema, but also may appear serosanguinous. The effusion is usually an exudate with increased protein content above 5 g/dl, with lymphocytic predominance in about 90% of cases, low or normal levels of pleural glucose and low pH values. Eosinophilic effusion with more than 10% in the pleural fluid is not common for tuberculous pleurisy and in these cases the differential diagnosis should be considered(2,4,6). Lactate dehydroge-nase (LDH) level in pleural fluid is elevated in tubercu-lous pleuritis, serum LDH levels being lower than pleural LDH levels. Mesothelial cells are also present in the tuberculous pleural fluid, but usually not more than 5%, the absence of these cells can’t be considered diagnostic for tuberculous pleuritis(4).
Tuberculin skin test
The tuberculin skin test is less useful in the diagno-sis of tuberculous pleurisy in patients with high diag-nostic suspicion, primarily because a negative skin test in these patients does not invalidate the diagnostic (low negative predictive value). In patients with severe immu-nosuppression (for example, a HIV-positive patient), the skin test may remain negative(4). In countries with a high prevalence of tuberculosis infection, a positive tubercu-lin skin test in a patient with an exudative pleural effu-sion may be useful for the positive diagnostic(4).
Interferon gamma release assays (IGRA)
This serum test has some diagnostic limits, and simi-lar to PPD skin test, it cannot distinguish latent from active tuberculosis. Quantiferon TB Gold and T Spot TB are the two forms of IGRA test available in Europe. Two MTB region antigen sequences, CFP10 and ESAT 6, are used to measure the quantitative IFN gamma response to specific sensitized lymphocytes(6).
A histopathological confirmation of tuberculous pleu-risy may be obtained performing pleural biopsy. The histo-pathological examination performed from biopsied pleural tissue is considered the gold standard for the diagnosis. This technique is recommended when clinical findings or pleural fluid examination provide ambiguous results, espe-cially for the differential diagnosis with malignancy. The HP pleural specimen can be obtained using biopsy tech
niques like blind pleural needle biopsy, imaging guided needle biopsy or thoracoscopy(6). The presence of acid fast bacilli is not mandatory if the histological examination provides the evidence of caseating granulomas(3).
Biomarkers and other pleural fluid assays
Low sensitivity of microbiological evaluation of pleu-ral fluid and the invasiveness of the pleural biopsies procedures have encouraged the researchers to study the accuracy of biomarkers in pleural fluid, thus obtaining a more rapid and less invasive diagnostic of tuberculous pleuritis(4). A lot of potential markers were evaluated in the last decades, but the majority of them have limited diagnostic accuracy(2).
Interferon-γ represents one of the most studied cytokines, because it has an important role in the immu-nologic response to mycobacterial infection(3). Interferon-γ is a proinflammatory cytokine produced by the Th1 cells, cytotoxic T cells and natural killer cells, and it helps the macrophages to increase their bacteri-cidal activity(4). An important diagnostic value of this marker was obtained by the measurement of the unstim-ulated pleural fluid IFN-γ concentration by ELISA method(3).
In the last decades, many studies were developed to demonstrate that elevated levels of IF-γ in the pleural effusions represents a reliable biomarker for tuberculous pleurisy and in the same time to establish its specificity and sensitivity. From the analyses of 13 studies, Greco et al. found the mean sensitivity and specificity of 87% and 97%, respectively(2,7). In another study, which included 145 cases of tuberculous pleurisy, an IFN-γ concentration more than 140 pg/mL had greater sensitivity, of 94%, and a specificity of 92%(3,8). Even if a lot of researches demon-strated its important diagnostic role, this biomarker isn’t yet widely available in clinical practice, because of its measurements cost, which is compared in one study with the cost necessary for the treatment of six patients. Of real interest for future researchers is to establish its posi-tive predictive value, because at this moment there are no information about it(3).
Adenosine deaminase (ADA)
ADA represents one of the most studied biomarkers from the pleural fluid and many researchers had focused their attention on it because is an inexpensive method to establish the tuberculous etiology of a pleural effu-sion(4). ADA is the enzyme that catalyzes the conversion of adenosine and deoxyadenosine to inosine and deox-yinozine, respectively. ADA is involved in the prolifera-tion and differentiation of lymphocytes, especially T-lymphocytes. The T-cells release ADA during the pro-cess of activation in the presence of live intracellular pathogens(5,6).
In 1978, the activity of ADA was detected to be increased in tuberculous pleural effusions. Increased pleural ADA levels can also be detected in pleural effusion secondary to other diseases as neoplasia, pneumo-nia, lymphoma and autoimmune diseases like systemic lupus erythematous(7.8,9). ADA levels may also be elevated in HIV-positive patients, but empyema remains, after TB, the second etiology with elevated ADA levels(4).
ADA is composed of two main isoenzymes, ADA1 and ADA2, which have different optimal pH and relative speci-ficity patterns. In humans, isoenzyme ADA1 is ubiquitous, being present in many cells. ADA2 is produced mainly by monocytes/macrophages and is responsible for the increased activity of ADA in tuberculous pleurisy(2).
The role of ADA in the diagnostic of tuberculous pleurisy diagnostic depends on the prevalence of the tuberculosis in different regions. In countries with high prevalence of tuberculosis, the sensitivity and specific-ity of this marker are high(5). Elevated levels of ADA in pleural effusions can be considered diagnostic for tuber-culous pleuritis and also might justify the initiation of specific treatment. Earlier test of pleural fluid ADA lev-els on the onset of the pleurisy demonstrated low levels of ADA, but performing repeated thoracentesis, elevated levels of ADA can be found. Very low ADA levels or nega-tive ADA test may justify abandoning further diagnostic tests for tuberculous pleurisy(2).
Many researchers tried to find a cut-off value of pleural fluid ADA, but as we previously specified, it depends on the prevalence of the tuberculosis in different regions. The most widely accepted cut-off value is 40 U/l(4,10).
Studies that followed the accuracy of different bio-markers in tuberculous pleural effusions showed the highest specificity and sensitivity for IFN-γ and total ADA levels. A previous meta-analysis that included 13 studied focused on the accuracy of IFN-γ and 31 which reviewed the ADA activity based on 1189 patients con-cluded that both biomarkers have high accuracy in diag-nosing TB pleurisy(4,11). Other studies that focused on directly comparing ADA and IFN-γ in patients with tuberculous pleurisy have reported slightly higher diag-nostic accuracy for IFN-γ(2,12,13). In clinical practice, both biomarkers are useful to avoid the invasive diagnostic procedures, the significant difference between test pric-es making ADA the most preferred diagnostic test(2,12).
Scoring systems based on combination of tests
Even though single biomarkers as pleural fluid ADA and IFN-γ levels demonstrated their high sensitivity and specificity, the diagnosis of pleural TB remains a chal-lenge, and a positive biomarker test can’t establish by itself the etiologic agent.
Due to the limitations of single tests, combinations of tests have been developed. Such a scoring system based on clinical and pleural fluid assays was created by Porcel and Vives(1,14). They created two scoring systems, one with and one without the measurement of ADA activity, and their purpose was to differentiate between tuberculous and malignant pleural effusions. In the first model, 4 parameters predicted the tuberculous etiology: ADA cut-off ≥40 U/l (five points), age <35 years old (two points), temperature ≥37.8 °C (two points) and pleural fluid red blood cell count <5*109 L-1 (one point). In the second one, TB was predicted by: age <35 (two points), temperature ≥37.8 °C (two points), pleural fluid red blood cell count <5*109 L-1 (one point) plus no history of malignancy (three points), pleural protein ≥50g L-1 (one point) and pleural fluid/serum LDH ratio ≥2.2 (one point). Summated scores demonstrate high sensitivity and specificity, more than 94% for distinguishing between TB and malignant effusions(1,14).
Another combination of tests was developed by Villegas et al. They used the combination of PCR, IFN-γ and ADA activity, and their results stated a higher speci-ficity and sensitivity compared with single methods(1,15).
In the future, the determination of such diagnostic scoring systems could facilitate the diagnostic of tuber-culous pleurisy, avoid expensive and time-consuming procedures and in the same time decrease the interval between symptoms onset and specific treatment(16).
Tuberculous pleurisy remains a subject of interest for researchers. The majority of the studies demonstrat-ed the important role of pleural fluid biomarkers in diagnosing tuberculous pleurisy. Pleural level of Interferon-γ and total ADA level have the best sensitiv-ity and specificity. Another useful and accurate test is represented by isoenzyme ADA2, but in many countries its cost makes it accessible only for researches.
Even though bacterial culture from pleural fluid and histopathological examination performed from biopsied pleural tissue are considered the gold standard for the diagnosis, rapid and useful scoring systems combining clinical and biological markers may represent a viable alternative diagnostic tool, but further work is neces-sary to identify the best one for the clinical practice.
- A. Trajman,M.Pay, K. Deda, R. van Zyl Smit, A.A. Zwerlling, R. Joshi, S. Kalantri, P. Daley, D. Menzies. Novel tests for diagnosing tuberculous pleural effusion: what works and what does not? Eur Respir J 2008; 31:1098-106.
- Light RW. Update on tuberculous pleural effusion. Respirology 2010; 14, 451-8.
- Doosoo Jeon, M.D. Tuberculous Pleurisy: An Update. Tuberc Respir Dis 2014; 76:153-9.
- Vorster MJ, Allwood BW, Diacon AH, Koegelenberg CF. Tuberculous pleural effusions: advances and controversies. J. Thorac Dis 2015:7(6); 981-91.
- J. Ferrer.Pleural tuberculosis. Eur Respir J 1997; 10:942-7.
- KG.Bharat, V Bharat, D Bandyopadhyay. Role of Adenosine Deaminase Estimation in Diferentiation of Tuberculous an Non-tuberculous Exudative Pleural Effusions, J Clin Med Res 2010; 2(2); 79-84.
- L. Valdes, E, San Jose, D Alvarez, J.M. Valle. Adenosine deaminase (ADA) isoenzyme analysis in pleural effusions: diagnostic role, and relevance to the origin of increased ADA in tuberculous pleurisy. Eur Respir J, 1996, 9, 747-51.
- J Ena, V Valls, C Perez de Otezya, R Enrique de Salamanca, Utilidad y limitations de la adenosina desaminasa an el diagnostic de la pleuresia tuberculosa; studio metanalitico. Med Clin (Barc) 1990; 95:333-5.
- C. Gakis. Adenosine deaminase (ADA) isoenzymes ADA1 and ADA2: diagnostic and biological role. Eur Respir J 1996, 9, 632-3.
- JM Porcel. Tuberculous pleural effusion. Lung 2009; 187:263-70.
- S Greco, E Girardi, R Masciangeloet al. Adenosine deaminase and interferon gamma measurements for the diagnosis of tuberculous pleurisy: a meta-analysis. Int. J. Lung Dis, 2003; 7:777-86.
- R Krenke, P Korczynski. Use of pleural fluid levels of adenosine deaminase and interferon gamma in the diagnosis of tuberculous pleuritis. Curr Opin Pulm Med 2010; 16:367-75.
- J Jiang, HZ Shi, QL Liang, XJ Qin. Diagnostic value of interferon gamma in tuberculous pleurisy: a metaanalysis. Chest 2007; 131:1133-41. 14. JM Porcel, M Vives. Differentiating tuberculous from malignant pleural effusions: a scoring model. Med Sci Monit 2003; 9:174-80. 15. MV Villegas, LA Lbrada, NG Saravia. Evaluation of polymerase chain reaction, adenosine deaminase and interferon-γ in pleural fluid for the differential diagnosis of pleural tuberculosis. Chest 2000; 118:1355-64.