AB0663 18f-fdg-pet/ct disease distribution in a large vessel vasculitiscohort – supports vascular ultrasound as a screening and diagnostic tool

Abstract

BackgroundSubsets of GCA have extensive vascular involvement, termed Large Vessel GCA (LV-GCA), seen in 12%–37% depending on imaging used. These patients have higher relapse rates and are more often refractory to glucocorticoids (GC), requiring steroid-sparing treatment to minimise GC toxicity and vascular complications. Diagnosis is reliant upon imaging, given the relative inaccessibility of tissue for histological analysis beyond temporal artery biopsy (TAB). If axillary and subclavian arteries are often affected, vascular ultrasound could be an effective screening tool.ObjectivesTo determine disease distribution in the Southend Large Vessel Vasculitis (LVV) cohort, particularly levels of axillary and subclavian involvement.MethodsThe cohort has sixty-five patients (2010–2017), from which sixty 18F-FDG-PET/CT scans were performed. They were retrospectively examined by two nuclear radiologists. Vascular involvement was determined by consensus opinion at six territories (thoracic-aorta, abdominal-aorta, subclavian, axillary, carotid, vertebral, and iliac and femoral). Six negative scans were excluded from analysis, with LVV confirmed from other evidence (ultrasound, computed tomography). Of these, four were on concurrent GC (7–12 mg prednisolone), raising the possibility of “false negatives”. Nineteen scans were positive despite GC (1–60 mg prednisolone). Higher doses tended to be of short duration, being unable to postpone due to clinical necessity, but exceeded the 3 day limit at which FDG signal starts to attenuate.Abstract AB0663 – Figure 1Disease distribution via 18F-FDG-PET/CTResults14.8% (n=8) had isolated supra-aortic disease, 14.8% (n=8) had isolated aortic disease, and 29.6% (n=16) had involvement in both (Fig 1). 72.2% of scans had involvement at either or both of the axillary and subclavian territories. Femoral and iliac involvement alone was seen in 1.9% (n=1), and in addition to aortic involvement in 7.4% (n=4). In this latter group, it was limited to the abdominal aorta in half, and included 1 case of Retroperitoneal Fibrosis. 31.5% (n=17) had involvement in all 3 regions. 24.1% (n=13) had no supra-aortic involvement.ConclusionsThis cohort demonstrates LVV has a predilection for aortic and supra-aortic regions. High axillary and subclavian involvement supports the use of vascular ultrasound as an effective imaging tool. Further imaging would still be warranted if suspicion remained high despite negative ultrasound, or to assess for vascular complications and alternate pathology. 18F-FDG-PET/CT is not without limitations. Atherosclerosis and vascular remodelling display increased FDG uptake, so requires cautious interpretation. Further research on GC influence is also needed. As such it has a role in monitoring flaring or non-responsive patients, rather than routine interval scanning. For the latter, ultrasound may be an alternative, especially if there is previously documented intimal-medial thickening at ultrasound accessible sites.References[1] Dejaco C, Brouwer E, Mason JC, Buttgereit F, Matteson...