Indexed on: 05 Mar '98Published on: 05 Mar '98Published in: The British journal of radiology
The aim of this study was to assess indium-111 leucocyte imaging and ultrasound in the investigation of pyrexia of unknown origin (PUO), as well as the predictive value of available clinical information on the radiological outcome. 256 111In leucocyte scans performed over a 5 year period were reviewed. There were 59 scans in 58 patients who fulfilled recognized criteria for true PUO; 47 of these patients had ultrasound. In 15 patients the fever settled and no diagnosis was made; the source of PUO was infective in 20 and non-infective in 24. In the group as a whole, the sensitivities of leucocyte scan and ultrasound were 25% and 23% with specificities of 100% and 83%, respectively. In infective cases of PUO, the sensitivities were 20% for both modalities. There was no correlation between leucocyte count or differential, C reactive protein or the presence or absence of antibiotics. In the cases where the leucocyte scan led to the diagnosis of infection, the scan was performed within 4 weeks of the onset of symptoms. The results show that an infective cause for PUO is established in less than 50% of cases. The leucocyte scan is specific and although not very sensitive can often be a guide to subsequent cross-sectional imaging. If an infective cause for the PUO is strongly suspected, 111In leucocyte scanning may be more sensitive if performed in the first few weeks of fever.