Diagnostic Yield of Computed Tomography-Guided Procedures for Spondylodiscitis

Purpose To identify the diagnostic yield and predictive factors for microbiological diagnosis in patients with spondylodiscitis through computed tomography (CT)-guided biopsy or aspiration. Materials and methods A retrospective review of 102 patients with suspected spondylodiscitis who underwent CT-guided procedures in an 8-year period was conducted. Analyzed variables were demographic data, C-reactive protein, pre-biopsy MRI findings, prior antibiotic use, site of biopsy/aspiration, histopathological findings, culture results and radiation dose. Results The culture yield among all sites was 56%, 85.6% for paravertebral/discal fluid aspiration, 66.7% for disk-only biopsy, 52.9% for paravertebral soft tissue biopsy, and 39.6% for endplate bone-disk unit. Patients with paravertebral/disk collection on MRI had better yields when submitted to fluid aspiration instead of biopsy of other sites (78.9% vs. 36.6%; p = 0,006). Pyogenic etiology corresponded to 68.3% of cases and Staphylococcus aureus was the most common agent. Prior antibiotics exposure was associated with a lower yield (66.2% vs. 40.9%, p = 0,016). Conclusions CT-guided procedures are safe and well-tolerated in patients with suspected spondylodiscitis, with good microbiologic yield particularly in the presence of paravertebral/discal abscess.