Introduction The aim of this study was to assess the accuracy of CT-guided core needle biopsy of musculo-skeletal tumours.

Methods This is a retrospective study on a series of 127 patients with a musculoskeletal tumours. The biopsies were performed over a four year period from 1998 to 2001. The accuracy of the CT-guided core needle biopsy was determined by comparing the histology of the biopsy with the final histology of the specimen obtained at open biopsy or surgical resection of the tumour. The effective accuracy was determined by the accuracy of the biopsy to diagnose benign versus malignant.

Results CT-guided core needle biopsy in this series has an overall accuracy of 80%. The effective accuracy as determined by a malignant versus benign lesion was 89%. There were 86 malignant tumours with a biopsy accuracy of 81% and there were 41 benign tumours with a biopsy accuracy of 78%. The positive predictive value (PPV) of a malignant tumour is 100% and the PPV of benign tumour 94.9%. The most common site of biopsy was from the femur and thigh, together accounting for 39.4% of the tumours. The most common tumours in this series were liposarcoma (n=12), osteosarcoma (n=11) and giant cell tumour (n=11). There were no reported complications arising from the biopsy.

Conclusions CT-guided core needle biopsy is a safe and effective procedure that is important in the diagnosis and management of musculoskeletal tumours.