Spinal tuberculosis: a longitudinal study with clinical, laboratory, and imaging outcomes.

Objectives: To study imaging changes associated with spinal tuberculosis (ST) over time and evaluate their association with clinical and laboratory data.

Methods: Between 1997 and 2004, patients with proven ST in our institution were prospectively enrolled and treated for 1 year. Clinical and laboratory data were collected at baseline and every 3 months. Computed tomography (CT) and magnetic resonance imaging (MRI) were performed at baseline. MRI was performed at least twice during treatment and at the end of treatment.

Results: Nineteen patients were enrolled; neurological signs were present at baseline in 14 patients (74%). Initial MRI/CT findings comprised (in percentages of patients) the following: paravertebral abscesses (100/100); intradiscal abscesses (47/0); and epidural abscesses (82/24) with spinal cord compression (64/18) or radicular compression (36/0). Edema of the vertebral body was observed on initial MRI in all the patients. Full follow-up data were available for 15 patients: all were cured with a mean follow-up of 25 months after the end of treatment. Weight gain was nearly maximal at 6 months, and pain relief was achieved within 9 months. C-reactive protein returned to normal after 3 months. On MRI, all epidural abscesses disappeared within 9 months; paravertebral abscesses disappeared after 3, 6, and 12 months in 45, 50, and 85% of patients, respectively. Vertebral body signal converted to a fatty signal in 75% of cases at 12 months.

Conclusions: Significant imaging abnormalities can persist after successful treatment of ST. These findings suggest that MRI need not be repeated in patients with clinical and laboratory improvement.