Validation of whole‐body against conventional magnetic resonance imaging for scoring acute inflammatory lesions in the sacroiliac joints of patients with spondylarthritis

Abstract

To compare the performance of whole-body magnetic resonance imaging (MRI) versus conventional MRI in assessing acute inflammatory lesions of the sacroiliac (SI) joints in patients with established and active spondylarthritis (SpA) using the Spondyloarthritis Research Consortium of Canada (SPARCC) MRI index. This study is validating whole-body MRI against the current MRI standard for assessing active inflammatory lesions of the SI joints in patients with SpA. Thirty-two SpA patients with clinically active disease (Bath Ankylosing Spondylitis Disease Activity Index score >/=4) fulfilling the modified New York criteria were scanned by whole-body and conventional MRI of the SI joints. The MRIs were scored independently in random order by 3 readers blinded to patient identity. Active inflammatory lesions of the SI joints were recorded on a Web-based SPARCC index. Pearson's correlation coefficients were used to compare scores for whole-body and conventional MRI for each reader, whereas intraclass correlation coefficients (ICCs) were used to compare interobserver reliability. The Pearson's correlation coefficients between whole-body and conventional MRI per rater were 0.94, 0.87, and 0.93. The mean sum scores for conventional versus whole-body MRI were statistically significantly higher for all 3 readers, although all patients showing inflammatory lesions on conventional MRI also demonstrated them on whole-body MRI. The ICCs(2,1) were 0.69, 0.78, and 0.95 for conventional MRI, and 0.79, 0.85, and 0.96 for whole-body MRI for the 3 possible reader pairs. Whole-body and conventional MRI scores show a strong correlation and comparable reliability for the detection of inflammatory lesions of the SI joints.