The differential diagnosis of the association of pneumonia and haemorrhagic pericarditis should include viral aetiologies

Extract

Sir,

For the sake of completeness, the differential diagnosis of bacterial pneumonia with haemorrhagic pericarditis¹ should include Chlamydia infection in immunocompetent subjects,² and the association of pneumonia and staphylococcal pericarditis.³ Where a bacterial pathogen has not been isolated,¹ and the atypical pneumonia screen, as is usually the case, does not extend beyond Mycoplasma pneumoniae, Legionella spp., Chlamydia pneumoniae and C. psittaci,⁴ the differential diagnosis also ought to include viral pathogens, given the association of viral pneumonia and haemorrhagic pericarditis in human rhinovirus type C infection,⁵ and in Coxsackie group B infection,⁶ respectively, and also in view of the association of haemorrhagic pleural effusion and haemorrhagic pericarditis in Coxsackie B infection.⁷ A favourable outcome after antibiotic treatment of what was perceived to be bacterial pneumonia with haemorrhagic pericarditis¹ does not necessarily validate presumption of a bacterial aetiology, given the fact that a similar outcome can occur when antibiotics are administered during the course of either rhinovirus type C pneumonia with pericarditis⁵ or Coxsackie B pleuropericarditis.⁷ Viral aetiology had been validated by real-time polymerase chain reaction targeting human rhinovirus in the first case,⁵ and by culture of pericardial fluid in the patient with Coxsackie B pleuropericarditis.⁷ In both instances, bacterial coinfection had been rigorously excluded by the fact that pericardial fluid was sterile on bacterial and mycobacterial culture. Where patients with bacterial pneumonia are coinfected with viral pathogens,⁸ there is also the potential risk that the viral agent might be the one that infects the pericardium, as was the case in an instance of Hemophilus influenza pneumonia where concomitant pericarditis was attributable to Coxsackie A9 cultured from the pericardial fluid.⁹ Even where polymorphs predominate in the pericardial fluid cell count this does not necessarily signify a bacterial aetiology given the fact that one example of Coxsackie pericarditis was characterized by a serosanguineous effusion with a total white cell count of 2100/mm³ including 1470/mm³ polymorphs.¹⁰