Suggestions for managing pyogenic (non-tuberculous) discitis in adults

Objectives: To develop recommendations about identifying the causative organism, obtaining imaging studies, and selecting pharmacological and non-pharmacological treatments in adults with pyogenic discitis and vertebral osteomyelitis (PDVO). Methods: A rheumatologist and an infectiologist drafted recommendations based on their personal experience and a review of studies in English or French retrieved on Medline using the following search terms: "infectious spondylodiscitis", "infectious spondylitis", "spondylodiscitis", "discitis", "vertebral osteomyelitis", "spine infection", and "bone and joint infections". The recommendations were submitted to four experts for validation. Results: 85 articles were selected for detailed review. No prospective randomized controlled trials were identified. Antimicrobial therapy should be initiated only after recovery of the causative organism in blood cultures or percutaneous disk biopsy specimens, except in patients with neutropenia or severe sepsis. The initial treatment rests on a combination of two bactericidal and synergistic antimicrobials in high dosages. The total duration of antimicrobial therapy should be 12 weeks at least. Radiographs of the spine and chest and magnetic resonance imaging (MRI) of the spine should be performed routinely during the initial evaluation. In PDVO due to hematogenous dissemination of a streptococcus or staphylococcus, routine echocardiography may be in order. Radiographs centered on the affected disk should be obtained 1 and 3 months into antimicrobial therapy and 3 months after treatment discontinuation. Follow-up MRI is usually unnecessary when the clinical and laboratory abnormalities respond to treatment. If not, or if the initial investigations show a collected abscess, a repeat MRI after I month of antimicrobial treatment may be useful. Clinical and laboratory follow-up is mandatory throughout antimicrobial therapy and during the first 6 months after treatment discontinuation. Conclusions: Recommendations based on descriptive studies and expert opinion were developed. They can be expected to improve the quality and uniformity of PDVO management. Further studies are needed to improve the level of evidence that is available for developing recommendations. In particular, prospective randomized multicenter studies should be performed to compare the intravenous to the oral route for initial antimicrobials administration and to compare different treatment durations. (C) 2007 Elsevier Masson SAS. All rights reserved.