The value of sacroiliac pain provocation tests in early active sacroiliitis

Abstract

Detection of preradiographic sacroiliitis is important for early diagnosis of ankylosing spondylitis (AS) and related spondyloarthropathies. Magnetic resonance imaging (MRI) is a valuable tool for the diagnosis of sacroiliitis in the early and active stages. The aim of this study is to assess the value of pain provocation tests in detecting early active sacroiliitis. Chronic low-back pain (LBP) patients were recruited and examined by blinded assessors for pain provocation tests: compression, distraction, Gaenslen, Mennel, Patrick, thigh thrust and sacral thrust tests. Patients underwent lumbar and sacroiliac MRI. The percentage of agreement for each pain provocation tests was between 72-95%, and the inter-rater reliability was from moderate to good (kappa, 0.43-0.87). Kappa values ranged from 0.43 to 0.60 with an agreement of 80-95% for clusters of pain provocation tests. As separately evaluated, pain provocation tests did not have favorable accuracy. When evaluated in clusters (out of three and five provocation tests) four positive over five tests on the left side reached an area under the curve 0.693 (95% CI 0.489-0.897), and two positive over three tests reached an AUC 0.697 (95% CI 0.484-0.910). Sacroiliac pain provocation tests had acceptable reliability in early active sacroiliitis; however, the discriminating capacity of these tests is poor. A multi-test regimen of three or five sacroiliac pain provocation tests may improve the accuracy of these tests discriminating sacroiliitis from LBP of mechanical origin. Four out of five selected tests or any of the two out of three selected tests have the highest predictive value.