Gribbons, Katherine

Objective To identify and validate, using computer‐driven methods, patterns of arterial disease in Takayasu arteritis (TAK) and giant cell arteritis (GCA). Methods Patients with TAK or GCA were studied from the Diagnostic and Classification Criteria for Vasculitis (DCVAS) cohort and a combined North American cohort. Case inclusion required evidence of large‐vessel involvement, defined as stenosis, occlusion, or aneurysm by angiography/ultrasonography, or increased 18 F‐fluorodeoxyglucose (FDG) uptake by positron emission tomography (PET) in at least 1 of 11 specified arterial territories. K‐means cluster analysis identified groups of patients based on the pattern of arterial involvement. Cluster groups were identified in the DCVAS cohort and independently validated in the North American cohort. Results A total of 1,068 patients were included (DCVAS cohort: TAK = 461, GCA = 217; North American cohort: TAK = 225, GCA = 165). Six distinct clusters of patients were identified in DCVAS and validated in the North American cohort. Patients with TAK were more likely to have disease in the abdominal vasculature, bilateral disease of the subclavian and carotid arteries, or focal disease limited to the left subclavian artery than GCA ( P < 0.01). Patients with GCA were more likely to have diffuse disease, involvement of bilateral axillary/subclavian arteries, or minimal disease without a definable pattern than TAK ( P < 0.01). Patients with TAK were more likely to have damage by angiography, and patients with GCA were more likely to have arterial FDG uptake by PET without associated vascular damage. Conclusion Arterial patterns of disease highlight both shared and divergent vascular patterns between TAK and GCA and should be incorporated into classification criteria for large‐vessel vasculitis.

Abstract Objectives To develop and replicate, using data-driven methods, a novel classification system in Takayasu’s arteritis based on distribution of arterial lesions. Methods Patients were included from four international cohorts at major academic centres: India (Christian Medical College Vellore); North America (National Institutes of Health, Vasculitis Clinical Research Consortium and Cleveland Clinic Foundation). All patients underwent whole-body angiography of the aorta and branch vessels, with categorization of arterial damage (stenosis, occlusion or aneurysm) in 13 territories. K-means cluster analysis was performed to identify subgroups of patients based on pattern of angiographic involvement. Cluster groups were identified in the Indian cohort and independently replicated in the North American cohorts. Results A total of 806 patients with Takayasu’s arteritis from India (n = 581) and North America (n = 225) were included. Three distinct clusters defined by arterial damage were identified in the Indian cohort and replicated in each of the North American cohorts. Patients in cluster one had significantly more disease in the abdominal aorta, renal and mesenteric arteries (P &lt; 0.01). Patients in cluster two had significantly more bilateral disease in the carotid and subclavian arteries (P &lt; 0.01). Compared with clusters one and two, patients in cluster three had asymmetric disease with fewer involved territories (P &lt; 0.01). Demographics, clinical symptoms and clinical outcomes differed by cluster. Conclusion This large study in Takayasu’s arteritis identified and replicated three novel subsets of patients based on patterns of arterial damage. Angiographic-based disease classification requires validation by demonstrating potential aetiological or prognostic implications.