Pericardial Tuberculosis – A Nightmare in Diagnosis

No comments

Diagnosing pulmonary tuberculosis can be a tricky thing. As highlighted previously, the Ziehl-Neelsen smear is not a terribly sensitive methodology and the MTB-PCR sensitivity depends on if the sample is smear positive. Given that pulmonary tuberculosis tends to have a higher bacterial burden, it would be reasonable to think that other diseases without this high burden would be difficult to diagnose. This is the case for tuberculous pericarditis.  This is because pericardial fluid is pauci-bacillary, with one study finding the diagnostic accuracy of culture being 50%, and stains ~5% (1). Other studies have cited that only 25-37% of effusions grow M. tuberculosis in culture (2). Because of these low percentages, and the fact that MTB culture tends to take weeks to grow, other diagnostic modalities have been implemented

Adenosine Deaminase

An early study (3) evaluated pericardial ADA level in 56 patients, divided into 5 groups: group I with tuberculous pericarditis as confirmed by culture/histology, group II with idiopathic pericarditis, group III with neoplastic pericarditis, group IV with either radiation or uremic pericarditis, and group V with no disease. Overall, patients with tuberculous pericarditis tended to have significantly higher levels of ADA:

Another study (4) evaluated 108 patients with pericardial effusion and classified patients into 5 groups (TB, idiopathic, neoplastic, purulent, and breast cancer). Diagnosis was made by histology or culture. TB pericarditis group had the highest mean ADA (126 +/- 16.68 U/). Idiopathic, neoplastic, purulent, and radiotherapy groups had 29-4 ±89 (min: 16, max: 74 U /1), 27 ±7-21 min: 21, max: 35 U/1), 29-5 ± 14 (min: 16, max: 39U/l) ,26 U/l  and 26 U/l respectively. All the tuberculous effusions had adenosine deaminase values of 70 U/l or more. With this value, sensitivity was 100% and specificity was 91%. Another prospective (5) study evaluated 233 patients with large pericardial effusions, using fluid culture/biopsy and histology as a comparison. After exclusion of those with anti-TB therapy and all those categorized as pericarditis without known cause, an ADA activity cut-off level of 40 U/l resulted in the corresponding sensitivity, specificity, PPV, NPV and diagnostic efficiency of 90.0%, 74.0%, 90.0%, 76.0% and 86.0%, respectively. Notably, however, is the fact that ADA levels were actually higher in those with septic pericarditis, and being on antituberculous therapy decreased the ADA levels significantly:

In a retrospective study (6) evaluating ADA in high incidence regions, the sensitivity and specificity of a 40 U/l as cut-off was 89% and 72%, respectively. Again, patients with septic pericarditis tended to have comparable ADA activity:

So while ADA is a reasonable test, septic pericarditis can also give you a high ADA.

GeneXpert MTB/Rif

So what about our friend, the GeneXpert MTB/Rif? The issue here is the fact that it depends on the smear positivity of the sample. As with pulmonary TB, the sensitivity of the GeneXpert Assay goes down with a negative smear. Remember that tuberculous pericarditis is a pauci-bacillary disease, so not a lot of bacilli to go around. So sensitivity is lower. In one study that evaluated the diagnostic utility of GeneXpert for extra-pulmonary tuberculosis (7), the sensitivity of smear positive specimens was 100% compared to 47.7% in the smear negative samples. This was still higher than the sensitivity for smear:

In another study evaluating GeneXpert to ADA and unstimulated IFN-=gamma (8), 151 patients were evaluated for pericardial tuberculosis. Overall, Xpert had a sensitivity of 63.8%, with ADA having a higher sensitivity with a cut-off of 35:

In a cross-sectional study (9), 839 extrapulmonary specimens were examined along with 1218 pulmonary specimens. Xpert had a sensitivity of roughly 40% in pericardial samples, which was one of the lower ones. Overall, in extra-pulmonary samples the sensitivity was 76.5%, lower than in pulmonary samples:

So GeneXpert may be useful, but not it is not perfect. One of the things to note is this data comes from areas of high incidence. The sensitivity of Xpert MTB/RIF in low incidence, high-resource settings is comparable in pulmonary specimens that are smear positive, but significantly lower in those that are smear negative. In one study, the sensitivity was found to be 86% (95% CI, 42%–100%) in the in smear-positive results, and 28% (95% CI, 10%–56%) smear-negative, culture-positive results(10). Another study from San Diego, CA in pulmonary specimens showed sensitivity was 89.6% (97.7% and 74.5% in smear-positive and -negative sputa, respectively) and specificity was 97.2%; while AFB smear sensitivity and specificity were 64.9% and 77.8%, respectively (11). In one study from France, the sensitivity of MTB-PCR in smear negative, culture positive extra pulmonary specimens was 37%, compared to pulmonary specimens (12).

 What about GeneXpert Ultra? Remember this assay detects MTB complex and it has multiple targets in comparison to the normal MTB/Rif assay, so its sensitivity is higher.

A retrospective study (13) evaluated the performance of the 2 assays in a tertiary care center. The sensitivity of pulmonary specimens was 87% and 75% in smear positive and smear negative samples, respectively. And in extrapulmonary specimens, this was 85% and 72% respectively.

So perhaps a bit better. Again, this depended on smear positivity. Having said that, there are other diagnostics that have been use for this. QuantiFERON TB-Gold test has been used in pleural specimens with a sensitivity of 74% and spec of 68%, lower than ADA (88% and 88% respectively (14). The utility of this has been unclear. Recently Mtb-specific CD4+ cells expressing CD38 and HLADR and Ki67 have been able to distinguish active and LTBI, however this is not ready for primetime yet.


  1. Reuter, H., et al. “Diagnosing Tuberculous Pericarditis.” Qjm, vol. 99, no. 12, Mar. 2006, pp. 827–839., doi:10.1093/qjmed/hcl123.Villegas, Maria Virginia, et al. “Evaluation of Polymerase Chain Reaction, Adenosine Deaminase, and Interferon-γ in Pleural Fluid for the Differential Diagnosis of Pleural Tuberculosis.” Chest, vol. 118, no. 5, 2000, pp. 1355–1364., doi:10.1378/chest.118.5.1355.
  2. Villegas, Maria Virginia, et al. “Evaluation of Polymerase Chain Reaction, Adenosine Deaminase, and Interferon-γ in Pleural Fluid for the Differential Diagnosis of Pleural Tuberculosis.” Chest, vol. 118, no. 5, 2000, pp. 1355–1364., doi:10.1378/chest.118.5.1355.
  3. Martinez-Vazquez JM, Ribera E, Ocaña I, Segura RM, Serrat R, Sagrista J. Adenosine deaminase activity in tuberculous pericarditis. Thorax. 1986;41(11):888-889. doi:10.1136/thx.41.11.888
  4. Komsuoğlu B, Göldelï O, Kulan K, Komsuoğlu SS. The diagnostic and prognostic value of adenosine deaminase in tuberculous pericarditis. Eur Heart J. 1995;16(8):1126-1130. doi:10.1093/oxfordjournals.eurheartj.a061057
  5. Reuter H, Burgess LJ, Carstens ME, Doubell AF. Adenosine deaminase activity–more than a diagnostic tool in tuberculous pericarditis. Cardiovasc J S Afr. 2005;16(3):143-147.
  6. 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):165-170. doi:10.1590/s0036-46652007000300006
  7. Zeka AN, Tasbakan S, Cavusoglu C. Evaluation of the GeneXpert MTB/RIF assay for rapid diagnosis of tuberculosis and detection of rifampin resistance in pulmonary and extrapulmonary specimens. J Clin Microbiol. 2011;49(12):4138-4141. doi:10.1128/JCM.05434-11
  8. Pandie S, Peter JG, Kerbelker ZS, et al. Diagnostic accuracy of quantitative PCR (Xpert MTB/RIF) for tuberculous pericarditis compared to adenosine deaminase and unstimulated interferon-γ in a high burden setting: a prospective study. BMC Med. 2014;12:101. Published 2014 Jun 18. doi:10.1186/1741-7015-12-101
  9. Allahyartorkaman M, Mirsaeidi M, Hamzehloo G, Amini S, Zakiloo M, Nasiri MJ. Low diagnostic accuracy of Xpert MTB/RIF assay for extrapulmonary tuberculosis: A multicenter surveillance. Sci Rep. 2019;9(1):18515. Published 2019 Dec 6. doi:10.1038/s41598-019-55112-y
  10.  Sohn, Hojoon, et al. “Xpert MTB/RIF Testing in a Low Tuberculosis Incidence, High-Resource Setting: Limitations in Accuracy and Clinical Impact.” Clinical Infectious Diseases, vol. 58, no. 7, 2014, pp. 970–976., doi:10.1093/cid/ciu022.
  11. Rice, Jason P., et al. “Performance of the Xpert MTB/RIF Assay for the Diagnosis of Pulmonary Tuberculosis and Rifampin Resistance in a Low-Incidence, High-Resource Setting.” Plos One, vol. 12, no. 10, Sept. 2017, doi:10.1371/journal.pone.0186139.
  12.  Armand, S., et al. “Comparison of the Xpert MTB/RIF Test with an IS6110-TaqMan Real-Time PCR Assay for Direct Detection of Mycobacterium Tuberculosis in Respiratory and Nonrespiratory Specimens.” Journal of Clinical Microbiology, vol. 49, no. 5, 2011, pp. 1772–1776., doi:10.1128/jcm.02157-10

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s