Automated interpretation (AI) systems for antinuclear antibody (ANA) analysis have been introduced based on assessment of indirect immunofluorescence (IIF) patterns. The diagnostic performance of a novel automated IIF reading system was compared with visual interpretation (VI) of IIF in daily clinical practice to evaluate the reduction of workload. ANA-IIF tests of consecutive serum samples from patients with suspected connective tissue disease were carried out using HEp-2 cells according to routine clinical care. AI was performed using a visual analyser (Zenit G-Sight, Menarini, Germany). Agreement rates between ANA results by AI and VI were calculated. Of the 336 samples investigated, VI yielded 205 (61%) negative, 42 (13%) ambiguous, and 89 (26%) positive results, whereas 82 (24%) were determined to be negative, 176 (52%) ambiguous, and 78 (24%) positive by AI. AI displayed a diagnostic accuracy of 175/336 samples (52%) with a kappa coefficient of 0.34 compared to VI being the gold standard. Solely relying on AI, with VI only performed for all ambiguous samples by AI, would have missed 1 of 89 (1%) positive results by VI and misclassified 2 of 205 (1%) negative results by VI as positive. The use of AI in daily clinical practice resulted only in a moderate reduction of the VI workload (82 of 336 samples: 24%).
Read full abstract