Percutaneous needle biopsy in diagnosis and identification of causative organisms in cases of suspected vertebral osteomyelitis

Background: Biopsy with demonstration of the infectious organism is the gold standard for diagnosing spondylodiscitis. The purpose of this study is to evaluate the positive culture rate of image-guided percutaneous biopsy in cases of radiologically suspected and unsuspected spinal osteomyelitis and to assess the role of pathology in diagnosis. Methods: With IRB approval and in compliance with HIPAA regulations, the charts of patients undergoing 323 consecutive image-guided percutaneous spinal biopsies performed by one musculoskeletal radiology department between January 2001 and March 2007 were reviewed. Image guidance was via fluoroscopy or computed tomography. Radiological and clinical suspicion, cultures, and pathology were assessed and compared to previously published reports. Results: In 92 cases radiographically and clinically consistent with infection (high probability of infection), 28 specimens yielded positive cultures (30.4%). Positive cultures resulted from 5 of 31 cases (16.1%) radiographically indeterminate for infection (intermediate probability of infection versus tumor). When radiographically not suggestive of infection (low probability of infection, i.e. suspicious for tumor), 10 of 200 cultures were positive (5.0%). From 113 cases sent to pathology with an intermediate or high suspicion for infection, 63 were histopathologically diagnosed as such (55.8%). Cultures were positive in 19 of those 63 cases (30.2%). Culture and/or pathology was positive in 73 (64.6%) of the 113 cases. There were no significant differences in rates of positive culture or pathology by vertebral region (p = 0.51, p = 0.81). The most frequently identified organisms were Staphylococcus aureus (13) and coagulase negative staphylococci (13). Conclusions: Our results suggest that the positive culture rate of percutaneous spinal biopsy specimens is 30.4% with radiographically high probability for infection, which is lower than previously published. Infection may also be present in cases with imaging characteristics atypical for infection (5.0%). Careful consideration must be given to the interpretation of negative culture results. Level of evidence: This retrospective review of 323 consecutive percutaneous spine biopsies is level III evidence.