The complexity of classifying ANCA-associated small-vessel vasculitis in actual clinical practice: data from a...

The different sets of criteria for diagnosis or classification of anti-neutrophil cytoplasmic antibody-associated vasculitis (AAV) lead to numerous overlapping and reclassified diagnoses in clinical practice. We designed this study to assess the difficulties in classifying patients with AAV. As a secondary objective, different variables were tested to predict prognosis. We conducted a retrospective chart review in a Western Spain multicentre survey. A total of 115 adult patients diagnosed with AAV from 2002 to 2013 and followed for at least 3 years were included. They were classified according to (1) Chapel Hill Consensus Conference (CHCC), (2) European Medicines Agency algorithm and (3) French Vasculitis Study Group/European Vasculitis Society phenotypes. Fifty-three patients (46%) had neither distinctive histopathological data of a single AAV definition nor any surrogate markers for granulomatous inflammation and thus did not fulfill any diagnostic criteria. Ocular, ear, nose, throat, skin, and lung involvement were more frequent with proteinase 3 (PR3) antibodies, whereas peripheral neuropathy was more frequent with myeloperoxidase (MPO) antibodies. When the disease was severe at diagnosis, the HR for mortality was 10.44. When induction treatment was not given in accordance with the guidelines, the HR for mortality was 4.00. For maintenance treatment, the HR was 5.49 for mortality and 2.48 for relapse. AAV classification is difficult because many patients had neither specific clinical data nor distinctive histological features of a single CHCC definition. A structured clinical assessment of patient severity is the best tool to guide the management of AAV.