The ability of follow-up imaging examinations to predict treatment failure in patients with spine infections has not been well studied.

We conducted a retrospective cohort analysis of patients with spine infection who had both baseline and 4-8-week follow-up imaging results available at the Mayo Clinic (Rochester, MN) during the period of 1994-2002. Follow-up imaging findings were categorized as improved, equivocal, or worse, compared with the baseline findings, on the basis of a simple grading system that focused on associated soft-tissue findings.

Baseline and 4-8-week follow-up imaging findings were available for 79 patients with spine infection who presented to the Mayo Clinic during 1994-2002. Thirty-five infections (44%) were due to Staphylococcus aureus, 9 (11%) were due to coagulase-negative staphylococci, and 16 (20%) were culture negative. Twenty-seven (34%), 38 (48%), and 14 (18%) follow-up images were graded improved, equivocal, or worse, respectively. The cumulative rates of 1-year survival free of microbiologically confirmed treatment failure were 100%, 89% (95% CI, 74%-96%), and 56% (95% CI, 24%-83%) for patients with improved, equivocal, and worse follow-up imaging findings, respectively (P=.004). Only 3 (6%) of 52 patients deemed to have had clinical improvement at the time of the follow-up imaging study experienced treatment failure. Elevated levels of inflammatory biomarkers identified 2 of these patients as high risk for treatment failure, and the levels were not performed for the third patient.

Applying a simple grading scale to assess follow-up imaging examinations for patients with spinal infection stratifies their risk of treatment failure. Patients' clinical statuses and inflammatory biomarker responses may be helpful for selecting patients at high risk for treatment failure who should undergo follow-up magnetic resonance imaging.