Magnetic resonance imaging examinations of the spine in patients with ankylosing spondylitis, before and after successful therapy with infliximab: Evaluation of a new scoring system

Abstract

Objective: To evaluate a magnetic resonance imaging (MRI) scoring system for the assessment of spinal inflammation in patients with ankylosing spondylitis (AS) who participated in a randomized, placebo‐controlled trial of infliximab, and to examine whether infliximab is also effective for the reduction of MRI‐proven spinal inflammation. Methods: Twenty patients with AS (9 women and 11 men, mean age 40.9 years) were examined at baseline and after 3 months. Nine patients had received infusions of infliximab (5 mg/kg body weight) at weeks 0, 2, and 6, and 11 patients had received placebo. Three MRI sequences and 2 scoring systems were used. Chronic lesions were evaluated by T1‐weighted turbo spin‐echo (TSE) sequences and were assigned a chronicity score. Active lesions were evaluated either by repetition of T1‐weighted TSE sequences after infusion of gadolinium–diethylenetriaminepentaacetic acid (Gd‐DTPA) or by short tau inversion recovery (STIR) sequences, and were assigned an activity score. The 40 images were evaluated twice by 2 readers who were blinded to the names of the patients and the dates of the examinations, and were analyzed in relation to the clinical results. Results: Active spinal lesions were detected in 15 of 20 patients (75%); the frequency as determined by STIR was equal in the 2 groups. At baseline, the total MRI scores determined using Gd‐DTPA, STIR, and T1 were 112.5, 156, and 253.5, respectively. The interrater variance and intrarater variance were, respectively, 6.4 and 7.7 for the active lesion score as determined by Gd‐DTPA, 15.7 and 5.3 for the active lesion score as determined by STIR sequence, and 167.3 and 75.5 for the chronic lesion score as determined by T1 sequence. Based on the means of the scores assigned by the 2 readers, the active lesion score as determined by Gd‐DTPA improved by 40% in the infliximab group compared with 6% in the placebo group, the active lesion score as determined by STIR improved by 60% in the infliximab group but deteriorated by 21% in the placebo group, and the chronic lesion score as determined by T1 improved by 7% in the infliximab group but worsened by 35% in the placebo group. Five patients in the infliximab group and 2 in the placebo group were clinical responders. The acute MRI changes correlated with clinical improvement as assessed by the Bath Ankylosing Spondylitis Disease Activity Index. Conclusion: This novel MRI scoring system performed well in assessing acute inflammation by using STIR and post–Gd‐DTPA sequences. In correlation with clinical improvement in patients with active AS who were treated with infliximab, significant regression of spinal inflammation was shown by using the MRI activity scores.