Background: The Birmingham Vasculitis Activity Score (BVAS) is a clinical assessment tool validated to assess disease activity in patients with systemic vasculitis. It allows clinical features to be represented as a numerical value which signifies disease activity. It has been widely used in clinical trials of vasculitis. A
modified version of the clinical tool (BVAS 2003) is currently being validated. BVAS has been used to screen for vasculitis in a limited study but its performance in patients with rheumatic presentations, not known to have vasculitis, has not been evaluated.
Methods: Patients not known to have systemic vasculitis but unwell due to other rheumatologic disorders were assessed using the BVAS and BVAS 2003. Both clinical tools require items to be scored only if they can be attributed to the presence of vasculitis. For this exercise, all items were scored regardless of attribution. We compared these findings with baseline data from patients with ANCA associated vasculitis who had been prospectively recruited into three large European Vasculitis Study group (EUVAS) clinical trials.
Results: 49 patients were recruited; 26 rheumatoid arthritis (RA), 4 reactive arthritis, 3 ankylosing spondylitis, 3 systemic lupus erythematosus (SLE), and 13 patients with other diagnoses. The mean scores for BVAS and BVAS 2003 were 5.54+/-�5.9 and 5.38�+/-6.24 respectively. 33/49 (67%) scored </=�5, and 40/49 (82%)
scored �</=10 (out of a maximum 63) on the BVAS. 35/49 (71%) scored �</=5 and 41/49 (84%) scored </=�10 on the BVAS 2003. BVAS and BVAS 2003 correlated well with each other (Spearman’s rank correlation coefficient �= 0.94). Reviewing the 9 patients with a BVAS >10, 2 of them had definite vasculitis (1 Wegener’s
granulomatosis, 1 rheumatoid vasculitis), 1 had probable vasculitis (WG), 2 patients had SLE and 4 patients did not have evidence of vasculitis. A BVAS>10 as a screening tool has a 33% chance of picking up occult vasculitis. 2/9 patients (1 SLE and 1 RA) who scored >10 on BVAS, died within 6 months of their assessment, supporting previous evidence of poor prognosis associated with high BVAS values in vasculitis. Comparing this data with the entry BVAS in 341 patients in EUVAS clinical trials, a BVAS�10 is 90% sensitive and 87% specific for a diagnosis of vasculitis.
Conclusions: A BVAS>10 is highly sensitive and specific for a diagnosis of vasculitis and can be used to screen for vasculitis in rheumatology patients. BVAS and BVAS 2003 scores correlated well in these patients.