The Diagnostic Performance of an Extended Ultrasound Protocol in Patients With Clinically Suspected Giant Cell Arteritis

Abstract

ObjectiveTo evaluate the diagnostic performance of an extended ultrasound protocol in patients referred under the suspicion of giant cell arteritis (GCA).MethodsConsecutive patients with suspected GCA were examined with an extended color duplex ultrasound (CDU) protocol during a period of 2 years. The extended CDU protocol included temporal, axillary, subclavian, brachiocephalic, and carotid arteries. The reference was clinically diagnosed GCA, confirmed after ≥6-month follow-up. Hypo- or medium-echogenic, circumferential, homogenous wall thickening, and/or a positive compression sign in temporal arteries, were regarded as typical signs of arteritis.ResultsOf the eligible 201 patients, 83 (41%) received a clinical GCA diagnosis at follow-up ≥6 months post CDU examination. Among these cases, 48 (58%) demonstrated inflammation solely in temporal arteries, 8 (10%) showed abnormalities restricted to extra-cranial vessels, and 23 (28%) patients displayed inflammatory changes in both temporal and extra-cranial arteries. Color duplex ultrasound of temporal arteries yielded a diagnostic sensitivity and specificity [95% confidence intervals (CI)] of 86% (76–92%) and 99% (95–99%), respectively. By adding axillary artery examination, the sensitivity increased to 92% (83–97%) while the specificity remained unchanged. Further, inclusion of subclavian artery marginally increased the sensitivity by 1%. Finally, by also including brachiocephalic and common carotid arteries resulted in a sensitivity of 95% (88–99%) and a specificity of 98% (94–99%).ConclusionsColor duplex ultrasound examination demonstrated a high accuracy in diagnosing patients both with cranial and extra-cranial GCA. Further examination of brachiocephalic and common carotid arteries can increase the sensitivity without affecting the specificity when temporal and axillary findings are indecisive. Finally, the extended CDU protocol allows measurement of the general burden of inflammation, which could be relevant for future monitoring purposes.