A. Dumont

Large-vessel involvement and aortic dilation in giant-cell arteritis. A multicenter study of 549 patients

OBJECTIVES Large-vessel involvement (LVI) can occur in giant-cell arteritis (GCA) and may represent a distinct disease subgroup with a higher risk for aortic dilation. This study aimed to better characterize the presentation and evolution of LVI in patients with GCA. PATIENTS AND METHODS A retrospective multicenter study enrolled 248 GCA patients with LVI and 301 GCA patients without LVI on imaging. Factors associated with aortic dilation were identified in a multivariable model. RESULTS The patients with LVI were younger (p<0.0001), more likely to be women (p=0.01), and showed fewer cephalic symptoms (p<0.0001) and polymyalgia rheumatica (p=0.001) but more extracranial vascular symptoms (p=0.05) than the patients without LVI. Glucocorticoids (GC) management did not differ between the two groups, but the GC discontinuation rate was lower in the patients with LVI (p=0.0003). Repeated aortic imaging procedures were performed at 19months [range: 5-162months] and 17months [range: 6-168months] after diagnosis in 154 patients with LVI and 123 patients without LVI, respectively, of whom 21% and 7%, respectively, presented new aortic dilations (p=0.0008). In the patients with LVI, aortic dilation occurred on an aorta segment shown to be inflammatory on previous imaging in 94% of patients. In the multivariate analysis, LVI was the strongest predictor of aortic dilation (hazard ratio: 3.16 [range: 1.34-7.48], p=0.009). CONCLUSIONS LVI represents a distinct disease pattern of GCA with an increased risk of aortic dilation. Control of the aortic morphology during follow-up is required.

Giant cell arteritis presenting as isolated inflammatory response and/or fever of unknown origin: a case-control study

The objective of this study was to determine the proportion and characteristics of patients with giant cell arteritis (GCA) who present with isolated inflammatory response and/or fever of unknown origin (IFUO). Using a cohort of 693 consecutive patients in two centers with evidence of GCA on biopsy and/or imaging, we compared the characteristics and outcomes of patients with IFUO at diagnosis to a control group made up of the remaining patients with GCA. Sixty-one (9%) patients initially presented with IFUO. GCA diagnosis was proven by biopsy in 50 (82%) patients and/or imaging in 23 out of 39 (59%) patients who underwent large-vessel imaging. At diagnosis, patients with IFUO were younger (p = 0.008), had longer time to diagnosis (p = 0.001), and showed more intense inflammatory response, i.e., had higher levels of C-reactive protein (p = 0.02) and lower hemoglobin levels (p = 0.0001) than control patients. However, the therapeutic regimen did not differ between the two groups. Similarly, during a median follow-up period of 50 [0–279] months, the total rate of cardiovascular events, including ischemic cranial complications and overall outcomes, including relapse, glucocorticoids-dependence and death rates did not differ between the two groups. Five (16%) patients with initial IFUO exhibited cranial symptoms at relapse. Giant cell arteritis presenting with isolated inflammatory response and/or fever of unknown origin is a well-defined demographic and clinical pattern affecting nearly 10% of patients. This clinical form is not associated with a particular prognosis but remains a challenging diagnosis.

AB0588 Concomitant association of giant cell arteritis and malignancy: a multicenter retrospective case-control study

Background Giant cell arteritis (GCA) is a large-vessel vasculitis affecting elderly people, and most frequently women (sex-ratio of 2.3). Some studies suggest an increased risk of malignancies in GCA. Objectives We aimed to describe the clinical, paraclinical characteristics and outcomes of GCA patients with concomitant malignancy and compare them to a control group without malignancy. Methods Patients with a diagnosis of GCA and of solid neoplasm or malignant blood disease, within one year before or after the diagnosis of vasculitis, were included. A random group of age-matched (3:1) control patients from our monocentric inception cohort of GCA patients from Caen University Hospital was constituted. Results Twenty-four observations were collected (median age 75.5 years). All fulfilled ≥3/5 ACR criteria. Temporal artery biopsy was positive in 17 cases (70.8%). There were 1 active (4.2%) and 9 former (37.5%) smokers. Only 1 patient had a previous prostate cancer. Malignancies were 10 malignant blood diseases (41.7%, 3 chronic lymphoid leukemias, 3 essential thrombocythemias, 1 myeloma, 1 chronic myelomonocytic leukemia, 1 MALT lymphoma, 1 Waldenströms macroglobulinemia) and 14 solid neoplasms (58.3%, 3 lung, 3 breast, 2 prostate, 1 thyroid, 1 colon, 1 pleural cancers, 1 melanoma, 1 Kaposis sarcoma and 1 Merkel cell carcinoma). Malignancy was diagnosed at a median of 1 month after GCA diagnosis in 21 patients and before in the other 3. Diagnosis of malignancy was made in consultation in 5 patients (3 skin cancers and 2 breast cancers), on lab tests in 13 (thrombocytosis, anemia or increased prostate specific antigen) and on imaging in 6. Treatments of malignancy included chemotherapy alone in 8 patients (33.3%), simple monitoring in 6 patients (25%), surgery alone in 4 patients (16.7%), surgery and radiotherapy and/or chemotherapy in 4 patients (16.7%), decrease of corticosteroids in 1 patient, and 1 patient was lost to follow-up. Two patients (8.3%) died from infectious complications, 8 patients (33.3%) had a GCA relapse, including one with concomitant malignancy relapse. After a median follow-up of 16 months [0–134], 5 patients (20.8%) were weaned from steroids, all considered in malignancy remission. Seven patients (29.1%) were still under chemotherapy, 9 patients (37.5%) were considered to be in malignancy remission. There were more males in patients with concomitant malignancy, compared to the control group (respectively 15/24 and 21/72, p<0.005). Conclusions Our study shows an over-representation of male gender in GCA with concomitant malignancy. Vasculitis outcomes were not influenced by the malignancy treatment. The diversity of malignancies encountered in this study raises the issue of an incidental association. Initial clinical and paraclinical follow-up dictated by vasculitis may have led to an early identification of associated malignancy, and thus represent a lead time bias. Disclosure of Interest None declared