Magnetic resonance imaging of the spine and the sacroiliac joints in ankylosing spondylitis and undifferentiated spondyloarthritis during treatment with etanercept

Abstract

Objective: To assess the changes in inflammatory lesions of the spine and the sacroiliac (SI) joints as detected by magnetic resonance imaging (MRI) in patients with ankylosing spondylitis (AS) and undifferentiated spondyloarthritis (uSpA) with predominant axial symptoms during treatment with etanercept.

Methods: MRI of the spine and/or the SI joints of patients with active AS or axial uSpA was performed at baseline (TP0, n = 25), after 6 weeks (TP1, n = 20), and after 24 weeks of continuous treatment with etanercept (TP2, n = 12). T1 weighted spin echo pre -(T1), post-gadolinium (T1/Gd-DTPA) and short tau inversion recovery (STIR) MRI sequences were used to assess chronic and active spinal lesions using the scoring system ASspiMRI. Active and chronic SI lesions were assessed using a simple scoring system.

Results: By use of the definite STIR sequence, significant regression of spinal inflammation was already seen already after 6 weeks in the patients treated with etanercept (mean (SD) 11.2 (13.8) at TP0 v 6.8 (7.9) at TP1; p = 0.023) but not in patients treated with placebo. Continuous treatment with etanercept for 24 weeks reduced active spinal changes by 69% (p = 0.012). T1/Gd-DTPA sequences gave similar results. There was only a trend for a decrease of active inflammatory lesions of the SI joints.

Conclusions: Etanercept treatment in patients with active AS and uSpA leads to regression of active inflammatory lesions of the spine as depicted by MRI. The potential role of etanercept on deceleration of chronic spinal changes needs further study.

Figure 1

Mean improvement of active spinal lesions as assessed by MRI using the STIR and T₁/Gd-DTPA technique and quantified by the ASspiMRI-a score at baseline and after 6 weeks of etanercept or placebo treatment, respectively. Significant differences were found for the patients treated with etanercept (in the STIR sequence and in the T₁/Gd-DTPA sequence) but not for patients treated with placebo.

Figure 2

Improvement of inflammatory spinal lesions in a patient with AS before and after application of T₁/Gd-DTPA. The hyperintensity of the signal in the vertebra is no longer detectable after 6 weeks of treatment with etanercept.

Figure 3

Active inflammatory lesions of the spine (n = 12) as assessed by MRI using the STIR sequence and as quantified by the ASspiMRI-a at baseline (TP0), after 6 weeks (TP1), and after 24 weeks of treatment (TP2). *p = 0.001 refers to the comparison between TP0 and TP2.

Figure 4

Active inflammatory lesions of the SI joints (n = 15) as assessed by MRI using the STIR sequence and a simple score at baseline (TP0), after 6 weeks of treatment (TP1), and after 24 weeks of treatment (TP2).

Figure 5

An example of a patient with significant improvement of active inflammatory lesions of the SI joints over the treatment period. Improvement was already seen after 6 weeks of treatment.