Validation of a multimodal algorithm for diagnosing giant cell arteritis with imaging

Purpose: The purpose of this study was to identify which combination of imaging modalities should be used to obtain the best diagnostic performance for the non-invasive diagnosis of giant cell arteritis (GCA). Materials and methods: This IRB-approved prospective single-center study enrolled participants presenting with a suspected diagnosis of GCA from December 2014 to October 2017. Participants underwent high-resolution 3T magnetic resonance imaging (MRI), temporal and extra-cranial arteries ultrasound and retinal angiography (RA), prior to temporal artery biopsy (TAB). Diagnostic accuracy of each imaging modality alone, then a combination of several imaging modalities, was evaluated. Several algorithms were constructed to test optimal combinations using McNemar test. Results: Forty-five participants (24 women, 21 men) with mean age of 75.4 +/- 16 (SD) years (range: 59-94 years) were enrolled; of these 43/45 (96%) had ophthalmological symptoms. Diagnosis of GCA was confirmed in 25/45 (56%) patients. Sensitivity and specificity of MRI, ultrasound and RA alone were 100% (25/25; 95% CI: 86-100) and 86% (19/22; 95% CI: 65-97), 88% (22/25; 95% CI: 69-97) and 84% (16/19; 95% CI: 60-97), 94% (15/16; 95% CI: 70-100) and 74% (14/19; 95% CI: 49-91), respectively. Sensitivity, specificity, positive predictive and negative predictive values ranged from 95 to 100% (95% CI: 77-100), 67 to 100% (95% CI: 38-100), 81 to 100% (95% CI: 61-100) and 91 to 100% (95% CI: 59-100) when combining several imaging tests, respectively. The diagnostic algorithm with the overall best diagnostic performance was the one starting with MRI, followed either by ultrasound or RA, yielding 100% sensitivity (22/22; 95% CI: 85-100%) 100% (15/15; 95% CI: 78-100) and 100% accuracy (37/37; 95% CI: 91-100). Conclusion: The use of MRI as the first imaging examination followed by either ultrasound or RA reaches high degrees of performance for the diagnosis of GCA and is recommended in daily practice. (C) 2021 Societe francaise de radiologie. Published by Elsevier Masson SAS. All rights reserved.