Effect of Nonsteroidal Anti-Inflammatory Drugs on Sacroiliac Joint Inflammation, as Seen on Magnetic Resonance Imaging, in Axial Spondyloarthritis.

OBJECTIVE

Imaging evidence of active sacroiliitis is important for diagnosis, classification, and monitoring of axial spondyloarthritis (axSpA). However, there is no consistent guidance on whether patients should temporarily stop nonsteroidal anti-inflammatory drugs (NSAIDs) before magnetic resonance imaging (MRI). The aim of this study was to determine whether NSAIDs lead to an underestimation of active sacroiliitis, as observed using MRI.

METHODS

Adults with axSpA were recruited from rheumatology clinics and undertook NSAID washout for one to two weeks before a sacroiliac joint MRI scan. Images were read by two independent readers and adjudicated by a third if required. Those who had a positive result for active sacroiliitis, as per internationally recognized criteria, underwent a second scan six weeks after recommencing daily NSAIDs. We determined the proportion of participants who had a negative scanning result while taking NSAIDs after a previous positive result when NSAID-free. Images were also scored using semiquantitative methods comprising lesion size and intensity, and a subset of participants underwent quantitative MRI (qMRI) to provide an objective evaluation of any inflammatory changes.

RESULTS

From 34 centers across the United Kingdom, 311 participants (median age 42 years; 62% male) were recruited; 286 (92%) completed the NSAID washout and underwent the first MRI scan. From 146 participants with active sacroiliitis, follow-up scans (while taking NSAIDs) were obtained from 124 (85%), at which point 25 participants had a negative result (20.2%; 95% confidence interval 13.5%-28.3%). Semiquantitative and qMRI methods supported these findings.

CONCLUSION

One-fifth of patients showed full resolution of active sacroiliitis lesions when NSAIDs were present. In clinical practice, if patients with axSpA are willing to attempt a one- to two-week NSAID washout before MRI, this should be considered.