ABSTRACT
STUDY DESIGN: A retrospective study of 103 computed tomography-guided biopsies of the spine. These represent a consecutive series of patients with spinal lesions or disorders observed over a 32-month period. OBJECTIVES: To determine the diagnostic accuracy and clinical usefulness of computed tomography-guided biopsies with respect to major influencing variables. SUMMARY OF BACKGROUND DATA: Computer tomographic-guided biopsy of the spine is considered a safe, accurate, and relatively inexpensive examination technique. A study comparing its diagnostic accuracy with respect to all the variables of age, gender, radiographic appearance, spinal level, tissue type, or pathologic diagnosis has not been done. METHODS: Biopsy specimens were sent for cytologic and histologic analysis. Bacteriologic studies were performed when clinically indicated. The biopsy results were analyzed for adequacy and diagnostic accuracy, i.e., the ability to generate a tissue sample adequate for pathologic examination and one that yields diagnostic information. RESULTS: The mean age of patients was 60 years, with a range of 4-91 years. The spines of 52 males and 51 females were studied. There were eight cervical, 28 thoracic, 53 lumbar, and 14 sacral lesions used as biopsy sites. The radiographic appearance of spinal lesions were lytic in 74 cases, blastic in four cases, and mixed in two cases. Tissues undergoing biopsy included bone (63 cases), soft tissue (35 cases), and mixed specimens (five cases). The pathologic examinations revealed 18 infections, 23 primary neoplasms, 34 metastases, and 19 normal tissues. An adequate specimen for pathologic examination was obtained in 90 biopsies (87%). A diagnosis was achieved in 67 of 94 patients (71%). Diagnostic rates obtained in thoracic level biopsies were lower than those from biopsies of other spinal levels (P = .007). CONCLUSION: Computed tomography-guided biopsy is an important tool in the evaluation of spinal lesions. A positive biopsy result may preclude the need for open surgical intervention. This study included one of the largest series of patients in the medical literature. In addition, it determined the diagnostic rates of this procedure with respect to the major influencing variables. Thoracic-level biopsies have a diagnostic rate that is significantly lower than that of other spinal levels. No significant correlation was found between diagnostic accuracy and age, gender, radiographic appearance, tissue type, or eventual diagnosis.
