Assessing the construct validity of clinical tests to identify sacroiliac joint inflammation in patients with non‐radiographic axial spondyloarthritis

Abstract

Magnetic resonance imaging (MRI) can be used to identify sacroiliac joint (SIJ) inflammation and provide an earlier diagnosis of nonradiographic axial spondyloarthritis (nrAxSpA). However, MRI is frequently a resource-limited examination. Our aim was to assess if a set of physical clinical tests can identify SIJ inflammation in patients with nrAxSpA. Twenty participants with nrAxSpA underwent two functional tests (active straight leg raise, and stork test on the support side) and four pain provocation tests (Gaenslen's, posterior pelvic pain provocation, Patrick's Faber and palpation of the long dorsal SIJ ligament) for the SIJ, and then proceeded to a contemporaneous reference standard MRI. The Spondyloarthritis Research Consortium of Canada scoring system (SPARCC) was used to score MRI. Specificity, sensitivity, and likelihood ratios (LR) were calculated for individual clinical tests, and for the composite of tests. Pain provocation tests were superior to functional tests, which showed poor accuracy. The Patrick's Faber test was the best performing procedure (sensitivity 71%, specificity 75%, positive LR 2.9, negative LR 0.4). When combining the provocation tests, a positive test in one out of two tests demonstrated the strongest predictive value (sensitivity 86%, specificity 62%, positive LR 2.2, negative LR 0.2). Sacroiliac joint pain provocation tests correlate modestly with inflammation. The Patrick's Faber test showed the greater LR to identify SIJ inflammation in patients with nrAxSpA. SIJ pain provocation tests may offer a simple and cost-effective way of identifying patients with nrAxSpA who are most likely to have MRI evidence of inflammation.