When and how to operate on spondylodiscitis: a report of 13 patients.

PURPOSE: Conflicting reports exist regarding the surgical indications, timing, approach, staged or not operation, and spinal instrumentation for patients with spondylodiscitis. Therefore, we performed this study to evaluate the outcome of a series of patients with spondylodiscitis aiming to answer when and how to operate on these patients.
MATERIALS AND METHODS: We retrospectively studied the files of 153 patients with spondylodiscitis treated at our institution from 2002 to 2012. The approach included MR imaging of the infected spine, isolation of the pathogen with blood cultures and/or biopsy, and further conservative or surgical treatment. The mean follow-up was 6 years (range 1-13 years). We evaluated the indications, timing (when), and methods (how) for surgical treatment, and the clinical outcome of these patients.
RESULTS: Orthopedic surgical treatment was necessary for 13 of the 153 patients (8.5 %). These were patients with low access to healthcare systems because of low socioeconomic status, third-country migrants, prisoners or intravenous drug use, patients in whom a bacterial isolate documentation was necessary, and patients with previous spinal operations. The most common pathogen was Mycobacterium tuberculosis. The surgical indications included deterioration of the neurological status (11 patients), need for bacterial isolate (10 patients), septicemia due to no response to antibiotics (five patients), and/or spinal instability (three patients). An anterior vertebral approach was more commonly used. Nine of the 13 patients had spinal instrumentation in the same setting. Improvement or recovery of the neurological status was observed postoperatively in all patients with preoperative neurological deficits. Postoperatively, two patients deceased from pulmonary infection and septicemia, and heart infarction. At the last follow-up, patients who were alive were asymptomatic; ten patients were neurologically intact, and one patient experienced paraparesis. Imaging showed spinal fusion, without evidence of recurrent spondylodiscitis. Complications related to the spinal instrumentation were not observed in the respective patients.
CONCLUSIONS: Conservative treatment is the standard for spondylodiscitis. Physicians should be alert for Mycobacterium tuberculosis spondylitis because of the low access to healthcare systems of patients with low social and economic status. Surgical indications include obtaining tissue sample for diagnosis, occurrence or progression of neurological symptoms, failure of conservative treatment, large anterior abscesses, and very extensive disease. Thorough debridement of infected tissue and spinal stability is paramount. The anterior approach provides direct access and improved exposure to the most commonly affected part of the spine. Spinal instrumentation is generally recommended for optimum spinal stability and fusion, without any implant-related complications.