Tailoring Diagnosis and Management of Pericardial Disease to the Epidemiological Setting

To the Editor: We read the comprehensive review “Pericardial Disease: Diagnosis and Management” by Khandaker et al¹ with interest. We would like to highlight certain aspects of tuberculous pericarditis that were not mentioned in the review but that we consider to be relevant to most patients with pericarditis who live in developing countries where tuberculosis (TB) is endemic. Developing countries harbor 80% of the population of the world. In countries that have a high burden of infection with human immunodeficiency virus (HIV), such as South Africa, tuberculous pericarditis accounts for 50% and greater than 90% of large pericardial effusions in HIV-negative and HIV-positive patients, respectively.² It is an important cause of cardiac morbidity and mortality and presents specific diagnostic and therapeutic challenges.^3,4

The tuberculin skin test is not a reliable test in this epidemiological setting because of the background prevalence of TB, mass Bacille Calmette-Guérin immunization, and the likelihood of cross-sensitization from environmental mycobacteria.⁵ In a contemporary series in South Africa, a positive tuberculin skin test result of 10 mm or greater had a sensitivity, specificity, positive predictive value, and negative predictive value of 89%, 56%, 82%, and 69%, respectively.⁶ This is relevant despite the high prevalence of TB as a cause of large pericardial effusions because, when the diagnosis is wrong in those treated on a presumptive basis, there is a 5-fold increased risk of death.⁴ A diagnosis can sometimes be established by demonstrating TB elsewhere, such as in the sputum, but these are insensitive methods. Even invasive methods to examine the pericardium or pericardial fluid directly are insensitive with respect to microscopic examination, culture, and polymerase chain reaction, resulting in the development of diagnostic algorithms that use tests of pericardial fluid such as assays for adenosine deaminase, interferon γ, or lysozyme, which have high sensitivities and specificities as indirect tests of TB.^3,6 At the University of Cape Town, routine tests of pericardial fluid include assays for adenosine deaminase, lactate dehydrogenase, and protein to differentiate between an exudative and transudative process, in addition to those listed by Khandaker et al.

Despite these efforts, the diagnosis sometimes remains a presumptive one. In areas with a high prevalence of TB, a pericardial effusion is often considered to be tuberculous in origin unless an alternative diagnosis is apparent. Furthermore, treatment often needs to be commenced before a bacteriological diagnosis is established. This approach does not apply in regions where the prevalence of TB is low. Therefore, the diagnostic and therapeutic approach to large pericardial effusion is dependent largely on the background prevalence of TB in the community of the patient.⁷