|Home | About | Journals | Submit | Contact Us | Français|
In a patient with pleural effusion, first-line investigations for tuberculosis, if negative, are not conclusive.
A Rwandan man aged 32 sought advice after two weeks of fever and malaise. He had returned to the UK three months previously from a visit to Uganda during which time he had not taken any malaria prophylaxis. On questioning, he said he had a dull ache in the left upper quadrant of his abdomen and lower back pain. His temperature was 38°C. There were no focal signs in the chest but a chest radiograph revealed a small left pleural effusion. An ultrasound of his abdomen was unremarkable. Haemoglobin was 13.8 g/dL, white cell count 6.7×109/L, erythrocyte sedimentation rate 63 mm/h. Two blood films for malaria parasites were reported negative. Broad spectrum antibiotics were started. The left pleural effusion was aspirated under ultrasound guidance but no organisms, including acid-fast bacilli, were identified or subsequently cultured. Polymerase chain reaction (PCR) amplification for Mycobacterium tuberculosis DNA on the pleural aspirate was also negative. A Heaf test was read after seven days as grade I and a HIV test was negative. Despite ten days of antibiotics the patient continued to have a swinging pyrexia. A chest radiograph revealed that the left pleural effusion now occupied the entire hemithorax. A pleural biopsy was taken and 1.6 L of straw-coloured fluid was obtained via a chest drain. The fluid contained glucose 1.3 mmol/L, lactate dehydrogenase 2163 IU/L, protein 49 g/L; no malignant cells were identified. A definitive diagnosis was obtained from the pleural biopsy (Figure 1). Several acid-fast bacilli were identified, and culture grew a fully sensitive M. tuberculosis. The patient was then started on isoniazid, rifampicin, pyrazinamide and ethambutol without incident.
Tuberculosis (TB) remains a global threat to human health,1 and this diagnosis was considered throughout the patient’s admission. However, several investigations for M. tuberculosis were negative. This is not unusual in pleural tuberculosis. Acid-fast bacilli were not identified in the pleural aspirate; but, unlike tuberculous empyemas, TB pleural effusions are probably generated by a delayed hypersensitivity reaction to mycobacterial antigens secreted into the pleural space rather than to the bacillus itself.2 The relative non-reactivity to the Heaf test is also not unusual, with negative tests reported in up to 31% of patients.3 Co-infection with HIV may also attenuate this test, but our patient was HIV-negative at the time of testing. PCR amplification assays for M. tuberculosis can provide a rapid diagnosis and estimation of drug sensitivities.4 However, reduced sensitivity in pleural disease is thought to be due to the presence of natural inhibitors of the PCR amplification process.4 Pleural biopsy with both histological and microbiological examination has been reported to have sensitivities of greater than 90% for this condition.5
In patients with pleural disease, the presence of negative first-line investigations for tuberculosis should not exclude this diagnosis. A pleural biopsy should be considered.