Role of Image-Guided Biopsy in the Diagnosis and Antimicrobial Management of Vertebral Osteomyelitis

Abstract

Purpose The aim of this study was to evaluate the role of fluoroscopy-guided vertebral biopsy in vertebral osteomyelitis and determine if it results in a change in antimicrobial management based on bone culture yield. Materials and Methods This study is a retrospective chart review of 83 patients (49 men, 34 women; age range, 28–92 years; mean age, 56.9 years) with a diagnosis of vertebral osteomyelitis on imaging (computed tomography or magnetic resonance imaging) who underwent a fluoroscopy-guided biopsy from January 2010 to May 2015. Data collection included patient demographic information; etiology/risk factors for vertebral osteomyelitis; antibiotic exposure within 30 days before biopsy; blood, urine, and bone culture results; and any postprocedural change in antibiotic therapy. The primary outcome is culture yield from biopsy, whereas the secondary outcome is to determine if a change in antimicrobial therapy occurred based on biopsy results. A change in antimicrobial management is defined as transitioning from empiric antibiotic coverage to narrow-spectrum antibiotics based on positive bone culture yield or discontinuation of empiric antibiotic coverage based on negative culture yield. Biopsy culture yield was tested against a change in antimicrobial management with statistical analysis including the Fisher exact test. Results Positive bone culture yields were obtained in 22 (27%) of the 83 cases. Of the 22 that were positive, only 11 resulted in a change in antimicrobial management where broad empiric coverage was switched to narrow-spectrum organism-specific coverage. Of the remaining 61 negative cultures, only 2 (3%) had a change in management where the empiric antibiotic treatment was discontinued. Whether the culture results were positive or negative, 70 patients (84%) did not have any change in antimicrobial management after biopsy with the results being statistically significant (odds ratio, 27.3; 95% confidence interval, 5.25–141.82; P < 0.0001). Conclusions Vertebral biopsies do not cause a significant change in antimicrobial management in the majority of cases.