Article Text

Download PDFPDF

A transient pleural effusion

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Q1: What are the possible causes of a spontaneously resolving pleural effusion?

The most common causes of spontaneous resolving pleural effusion are given in box 1.1

Box 1: Common causes of spontaneous resolving pleural effusion


  • Tuberculosis.

  • Parapneumonic effusion (non-empyemic).

    Connective tissue disease

  • Systemic lupus erythematosus.

  • Rheumatoid arthritis.


  • Heart failure.

  • Pulmonary infarction.

    Drug induced (rare)

  • Bromocriptine.

  • Dantrolene sodium.


Q2: Name one radiological and one diagnostic procedure that would help in making a diagnosis

As part of routine investigation a chest radiograph would be the simplest radiological investigation to exclude pulmonary pathology as a cause of his acute presentation. His chest radiograph showed miliary shadowing but no other abnormality (fig 1).

Figure 1

Close-up of chest radiograph taken during the acute presentation showing the area above the left diaphragm.

In view of the headache with vomiting as one of the main complaints a lumbar puncture should be done after exclusion of raised intracranial pressure. Computed tomography of the brain was normal and cerebrospinal fluid analysis from a lumbar puncture showed the following:

  • 55 white cells/ml with 70% monocytes.

  • Glucose 1.6 mmol/l (with a blood glucose 5.7 mmol/l).

  • Gram stain and Ziehl-Neelsen stain for acid-fast and alcohol-fast bacilli were negative.

Q3: What is the most likely cause of his acute illness and how would you treat him?

The two above mentioned investigations and the clinical course during his acute illness support the diagnosis of miliary tuberculosis with meningeal involvement. He was treated with isoniazid, rifampicin, ethambutol, and pyrazinamide and improved slowly over the next month. His cerebrospinal fluid grew Mycobacterium tuberculosis five weeks after the lumbar puncture, fully sensitive to all four antituberculosis drugs prescribed. A repeat chest radiograph after four months was normal and he returned to work as a meat factory worker.


Although no further investigations were undertaken during the initial presentation with a pleural effusion, tuberculosis was the most likely cause. Pleural effusion can be a manifestation of primary and post-primary tuberculosis. Primary tuberculous effusion occurs most frequently in young adults.2 The effusion is usually unilateral and of moderate size. The diagnosis is best made by pleural biopsy specimens, which show pleural granulomata and positive cultures in 60%–80% of patients with proved tuberculosis and can be improved by using multiple pleural biopsy specimens.3 Enhanced culture yield can be achieved by using bedside inoculation of pleural fluid.4 Routine analysis of pleural fluid for glucose, protein, and acid-fast organisms are of little help in making a diagnosis. The value of adenosine deaminase levels in pleural aspirate is controversial. The prognosis of patients with tuberculous pleural effusion depends on the treatment chosen. Simple bed rest commonly results in complete absorption of the effusion but the incidence of subsequent pulmonary or extrapulmonary tuberculosis is high. This is a particular problem in patients with immunosuppression. Treatment with standard antituberculosis drugs is indicated in all cases and reduces the subsequent development of pulmonary or extrapulmonary tuberculosis.5 In view of the increasing incidence of tuberculosis in Great Britain all attempts should be made to diagnose and treat patients with tuberculous pleural effusion.

Final diagnosis

Spontaneously resolving tuberculous pleural effusion with subsequent miliary tuberculosis.


I thank Dr J I G Strang for constructive comments.