Automated interpretation of ANCA patterns - a new approach in the serology of ANCA-associated vasculitis

• DOI: 10.1186/ar4119
• Publication Date: 2012-12-14T13:14:24Z
• Authors: Ilka Knütter; Rico Hiemann; Therese Brumma; Thomas Büttner; Kai Großmann; Marco Cusini; Francesca Pregnolato; Maria Orietta Borghi; Ursula Anderer; Karsten Conrad; Dirk Reinhold; Dirk Roggenbuck; Elena Csernok
• Keywords: ANCA-Associated Vasculitis

Abstract Introduction Indirect immunofluorescence (IIF) employing ethanol-fixed neutrophils (ethN) is still the method of choice for assessing antineutrophil cytoplasmic antibodies (ANCA) in ANCA-associated vasculitides (AAV). However, conventional fluorescence microscopy subjective and prone to high variability. The objective this study was evaluate novel pattern recognition algorithms standardized automated interpretation ANCA patterns. Methods Seventy ANCA-positive samples (20 antimyeloperoxidase ANCA, 50 antiproteinase3 ANCA) 100 controls from healthy individuals analyzed on ethN formalin-fixed (formN) by IIF were used as a 'training set' development algorithms. Sera 342 patients ('test set') with AAV other systemic rheumatic infectious diseases tested patterns using microscopy. Results Interpretation software developed enabling positive/negative discrimination classification (C-ANCA) perinuclear (P-ANCA). Comparison visual reading 'test revealed Cohen's kappa (κ) values 0.955 0.929 formN discrimination. Analysis regard between C-ANCA P-ANCA showed agreement (κ = 0.746) 0.847). There no significant difference regarding formN, well ( P > 0.05, respectively). Conclusions Pattern can assist IIF. Automated demonstrated consistency assessment.