• 
• 

Three Part Question

In [patients with tuberculous pericardial effusion] is [adenosine deaminase (ADA)] useful in [the diagnosis of tuberculosis]?

Clinical Scenario

A 65-year-old male, with past medical history of pulmonary tuberculosis, attends the Emergency Department with shortness of breath and central chest pain for 7 days. Physical examination reveals raised JVP and chest x-ray shows a globular-shaped heart. You wonder whether pericardial fluid adenosine deaminase level would assist in the diagnosis of tuberculous pericardial effusion.

Search Strategy

Medline, 1950–October 2009, using the Ovid interface.
(exp Adenosine Deaminase/ OR (adenosine deaminase or ADA).mp) AND (exp Tuberculosis, Cardiovascular/ OR ((exp Tuberculosis/ OR (TB or tubercul$).mp) AND (exp Pericarditis/ OR pericardi$.mp)))
LIMIT to human and english language

Search Outcome

34 papers were found, of which 21 were irrelevant. Five relevant papers were included in the systemic review [Tuon 2006]. This review and the remaining seven relevant papers are summarized in the table.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Tuon et al. 2007 Brazil	9 cases of tuberculous pericarditis and 39 cases with pericarditis other than tuberculous (12 neoplastic, 11 septic and of 16 unknown origin).	Retrospective case-control study (level 4).	Sensitivity and specificity for ADA cut-off of 40 U/L	Sensitivity 89%, specificity 72%	Retrospective analysis only.
Reuter et al. 2006 South Africa	Consecutive patients (n = 233) presenting with pericardial effusions underwent a predetermined diagnostic investigation	Prospective observational study (level 5).	Sensitivity, specificity, PPV and NPV for ADA cut-off of 40 U/L	Sensitivity 87%, specificity 89%, PPV 95%, NPV 72%	Small patient numbers. Single arm with no control. Interferon-gamma actually performed better as a test than ADA. These patients may be the same 233 as those included in a study published in 2005, which was included in Tuon et al 2006.
Tuon et al. 2006 Brazil	Five studies were selected with a final number of 462 patients.	Systematic review covering the period 1980-2005 (level 1b).	Sensitivity, specificity, PPV and NPV for ADA cut-off of 40 U/L	Sensitivity 88%, specificity 83%, PPV 83%, NPV 88%	Excluded papers with cut-off limit for ADA other than 40 U/L. Excluded non-English papers.
Cubero et al. 2006 Spain	83 patients with a diagnosis of pericardial effusion.	Prospective study (level 2).	Sensitivity, specificity, PPV and NPV for ADA cut-off of 60 U/L	Sensitivity 100%, specificity 90%, PPV 90%, NPV 100%	Only small number of patients was recruited. High level of ADA was used.
Burgess et al. 2002 South Africa	110 consecutive patients with large pericardial effusions undergoing pericardial aspiration.	Prospective study (level 2).	Sensitivity, specificity, PPV and NPV for ADA cut-off of 30 UL	Sensitivity 94%, specificity 68%, PPV 80%, NPV 89%	Low level of ADA was used. Interferon-gamma performed better than ADA. These patients may form part of a cohort reported in 2005 (and included in the systematic review by Tuon et al) and in the study by Reuter et al in 2006.
Aggeli et al. 2000 Greece	41 patients (age range 17–77 years) with significant pericardial effusion were included; diagnostic pericardiocentesis and pericardial biopsy was performed while serum and pericardial fluid ADA was measured.	Prospective study (level 2).	Sensitivity, specificity, PPV and NPV for ADA cut-off of 72 U/L	Sensitivity 100%, specificity 94%, PPV 87.5%, NPV 100%	Only small number of patients was recruited. A high level of ADA was used.
Komsuoglu et al. 1995 Lebanon	Deaminase activity measured in the pericardial fluid of 108 patients who initially of undetermined origin.	Prospective study (level 2).	Sensitivity and specificity	Sensitivity 100%, specificity 91%	The methods of definitive diagnosis are not clearly stated. Small number of patients recruited. No cut-off level of ADA stated.
Koh et al. 1994 Korea	26 patients with moderate to large pericardial effusion and 19 controls.	Cohort study (level 2).	Sensitivity and specificity for ADA cut-off of 40 U/L	Sensitivity 93%, specificity 97%	Only small number of patients was recruited. No PPV and NPV result.

Comment(s)

In areas where tuberculosis is endemic, it is not uncommon for patients with tuberculous pericardial effusion to present to the Emergency Department with shortness of breath and chest pain. While the diagnosis of pericardial effusion can be made based on clinical findings and echocardiography, the tuberculous cause of pericardial effusion is difficult to rule out at the initial consultation, as the definitive diagnosis of tuberculous pericardial effusion requires isolation of the tubercle bacillus from pericardial fluid, and this is often difficult and time-consuming. The role of ADA to facilitate early diagnosis of tuberculous pericardial effusion is supported as a readily available, efficient and inexpensive test. The cut-off value of 40 U/L is generally used with acceptable sensitivity and specificity. Further studies are needed to clearly define the optimum cut-off level of ADA in order to achieve the best clinical outcome. However, ADA is not the only possible marker for pericardial TB. Burgess et al. (2002) studied pericardial IFN-γ in 30 of their 110 subjects. With an IFN-γ cut-off set at 200 pg/L there was a 100% sensitivity and specificity for pericardial TB. The larger series reported in 2006 (Reuter et al.) used an IFN-γ cut-off of 50 pg/ml. This gave a sensitivity of 92% and a specificity of 100% for pericardial TB. This group also developed a diagnostic model using HIV status, leucocyte count, the lymphocyte/neutrophil ratio and either pericardial ADA or IFN-γ. The model using ADA ≥40 U/L had 96% sensitivity and 97% specificity for pericardial TB. When IFN-γ replaced ADA the model performed even better with a sensitivity of 98% and a specificity of 100% for the diagnosis.

Clinical Bottom Line

The present evidence confirms the clinical value of ADA activity as an adjunctive diagnostic marker for tuberculous pericardial effusion, but IFN-γ may be more useful.

References

1. Tuon FF, Silva VI, Almeida GM, Antonangelo L, Ho YL . The usefulness of adenosine deaminase in the diagnosis of tuberculous pericarditis. Rev. Inst. Med. Trop. Sao Paulo 2007; 49(3): 165-70.
2. Reuter H, Burgess L, van Vuuren W, Doubell A. Diagnosing tuberculous pericarditis. Quarterly Journal of Medicine 2006; 99(12): 827-39.
3. Tuon FF, Litvoc MN, Lopes MI. Adenosine deaminase and tuberculous pericarditis--a systematic review with meta-analysis Acta Trop. 2006; 99(1): 67-74.
4. Cubero GI, Rubin J, Martín M, Rondan J, Simarro E. Pericardial effusion: clinical and analytical parameters clue. Int. J. Cardiol. 2006; 108(3): 404-5.
5. Burgess LJ, Reuter H, Carstens ME, Taljaard JJ, Doubell AF. The use of adenosine deaminase and interferon-gamma as diagnostic tools for tuberculous pericarditis. Chest 2002; 122(3): 900-5.
6. Aggeli C, Pitsavos C, Brili S, Hasapis D, Frogoudaki A, Stefanadis C, Toutouzas P. Relevance of adenosine deaminase and lysozyme measurements in the diagnosis of tuberculous pericarditis. Cardiology 2000; 94(2): 81-5.
7. Komsuoglu B, Goldeli O, Kulan K, Komsuoglu SS. The diagnostic and prognostic value of adenosine deaminase in tuberculous pericarditis. Eur. Heart J. 1995; 16(8): 1126-30.
8. Koh KK, Kim EJ, Cho CH, Choi MJ, Cho SK, Kim SS, Kim MH, Lee CJ, Jin SH, Kim JM. Adenosine deaminase and carcinoembryonic antigen in pericardial effusion diagnosis, especially in suspected tuberculous pericarditis. Circulation 1994; 89(6): 2728-35.