CT-guided needle biopsy for musculoskeletal lesions

We determined the diagnostic accuracy rate of 207 cases that underwent CT-guided needle biopsy for musculoskeletal lesions during the past 10 years, and describe the efficacy and indications of this method. We retrospectively analyzed a consecutive series of 207 cases that presented to our oncology group and underwent CT-guided needle biopsy during the 10-year period between April 1998 and March 2008. Diagnostic accuracy was assessed statistically by anatomical location and final diagnosis. The biopsy site was the spine/sacrum in 70 cases, pelvis in 53, extremities in 51, rib/scapula in 20, and retroperitoneum in 13. Bone lesions were detected in 176 cases and soft tissue lesions in 31. The final diagnosis was metastatic tumor in 63 cases, primary bone tumor in 63, primary soft tissue tumor in 23, infection in 18, and hematopoietic malignancy in 16 and non-tumorous lesions in 24. The diagnostic accuracy rate for all cases was 90%. No serious complications were seen. No statistically significant difference was found by anatomical site. According to the final diagnosis, benign tumors/malignant tumors (p < 0.005), primary bone tumor (p < 0.01), and infectious conditions (p < 0.001) showed significantly low diagnostic accuracy rates. CT-guided needle biopsy is a safe method with high diagnostic accuracy for musculoskeletal lesions. However, in patients with a primary bone tumor, it should be indicated for the initial diagnosis. If a comprehensive assessment based on the imaging conflicts with the pathological findings, an open biopsy should be considered.