Influence of clinical information on the diagnostic validity of MRI in the detection of abacterial sacroiliitis

Purpose: to evaluate the influence of clinical information on the diagnostic validity of MRI in the detection of abacterial sacroiliitis. Evaluation of sensitivities, specificities, negative (NPV) and positive predictive values (PPV) for MRI with and without clinical information in comparison to a prior defined standard. Materials and methods: the results of routine MRI reports of 65 patients with the clinical suspicion of abacterial sacroiliitis referred by rheumatologists for MR assessment of the sacroiliac joints were retrospectively reviewed. These results were compared to the results of a blinded reading of the same examinations performed without any clinical information. The MRI protocol included T1-SE, STIR and T1-SE contrast-enhanced sequences with fat saturation. All patients were followed for at least four years. The standard of reference was defined by an experienced rheumatologist and included baseline and follow-up data (clinical, laboratory and imaging). Abacterial sacroiliitis was diagnosed in 19 patients. Results: under routine clinical conditions, MRI revealed a sensitivity of 94.7% and specificity of 97.8%. A PPV of 94.7% and a NPV of 97.8% were achieved. Without clinical information, MRI revealed a sensitivity and specificity of 64 and 80%, and a PPV of 49% and a NPV of 86%, respectively. Conclusion: the additional knowledge of clinical information for the MR diagnosis of abacterial sacroiliitis increases its diagnostic value for the assessment of abacterial sacroiliitis. (orig.)