The present study attempted to define the clinical? radiological, immunological and pathological characteristics of microscopic polyangiitis (MPA), and to separate them from classic PAN (c-PAN) and Churg-Stauss syndrome (CSS). In most cases, patients presenting microaneurysms and/or multiple vessel stenoses, which reflect medium-sized vessel involvement, did not have antineutrophil cytoplasmic antibodies (ANCA) (6.6%). Conversely. patients with glomerulonephritis almost never had abnormal angiograms. Furthermore, the clinical characteristics of ANCA-positive patients also indicate small-sized vessel involvement. Skin involvement (73.1 vs 26.7%, P less than or equal to 0.05), glomerulonephritis (35.5 vs 0%, P less than or equal to 0.001) and the presence of ANCA (34.6 rs 6.7%, P less than or equal to 0.05 were significantly more frequent in patients with normal than abnormal angiograms, respectively. Conversely, hypertension (66.7 rs 23.1%, P less than or equal to 0.02), renal vasculitis (46.7 rs 0%, P less than or equal to 0.001) and hepatitis B antigenaemia (60 rls 11.5%, P less than or equal to 0.01) were significantly more common in patients with abnormal angiograms. Stratification of patients according to vessel size showed that, except for skin involvement (P less than or equal to 0.05)) and glomerulonephritis (P less than or equal to 0.01), which are direct manifestations of small-sized vessel diseases, clinical symptoms of PAN ol CSS, angiographic findings and ANCA were nut correlated to arteriole size. Although at present it is not possible to separate definitively MPA from c-PAN. our results show that ANCA should be considered diagnostic for MPA and, in most cases, should be an exclusion criterion for e-PAN. Conversely, small-sized vessel involvement call be observed in patients presenting characteristics of c-PAN, MPA or CSS and, therefore, is not a sufficient criterion for assigning diagnosis.