Comparison of magnetic resonance angiography and¹⁸F-fluorodeoxyglucose positron emission tomography in large-vessel vasculitis

<jats:sec><jats:title>Objectives</jats:title><jats:p>To assess agreement between interpretation of magnetic resonance angiography (MRA) and<jats:sup>18</jats:sup>F-fluorodeoxyglucose positron emission tomography (PET) for disease extent and disease activity in large-vessel vasculitis (LVV) and determine associations between imaging and clinical assessments.</jats:p></jats:sec><jats:sec><jats:title>Methods</jats:title><jats:p>Patients with giant cell arteritis (GCA), Takayasu’s arteritis (TAK) and comparators were recruited into a prospective, observational cohort. Imaging and clinical assessments were performed concurrently, blinded to each other. Agreement was assessed by per cent agreement, Cohen’s kappa and McNemar’s test. Multivariable logistic regression identified MRA features associated with PET scan activity.</jats:p></jats:sec><jats:sec><jats:title>Results</jats:title><jats:p>Eighty-four patients (GCA=35; TAK=30; comparator=19) contributed 133 paired studies. Agreement for disease extent between MRA and PET was 580 out of 966 (60%) arterial territories with Cohen’s kappa=0.22. Of 386 territories with disagreement, MRA demonstrated disease in more territories than PET (304vs82, p<0.01). Agreement for disease activity between MRA and PET was 90 studies (68%) with Cohen’s kappa=0.30. In studies with disagreement, MRA demonstrated activity in 23 studies and PET in 20 studies (p=0.76). Oedema and wall thickness on MRA were independently associated with PET scan activity. Clinical status was associated with disease activity by PET (p<0.01) but not MRA (p=0.70), yet 35/69 (51%) patients with LVV in clinical remission had active disease by both MRA and PET.</jats:p></jats:sec><jats:sec><jats:title>Conclusions</jats:title><jats:p>In assessment of LVV, MRA and PET contribute unique and complementary information. MRA better captures disease extent, and PET scan is better suited to assess vascular activity. Clinical and imaging-based assessments often do not correlate over the disease course in LVV.</jats:p></jats:sec><jats:sec><jats:title>Trial registration number</jats:title><jats:p><jats:ext-link xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="NCT02257866" ext-link-type="clintrialgov" specific-use="clinicaltrial">NCT02257866</jats:ext-link>.</jats:p></jats:sec>