Association of atypical anti-neutrophil cytoplasmic antibody with comorbidities and outcome in a hospital-based population.

Introduction: Atypical anti-neutrophil cytoplasmic antibody (a-ANCA) is characterized by a positive fluorescence staining other than typical cytoplasmic or perinuclear ANCA. ANCA is associated with increased risk of dialysis and mortality in patients with ANCA vasculitis. However, comorbidities related to a-ANCA and whether a-ANCA exhibits an increased risk for renal failure and mortality remain unclear. This study aimed to explore the comorbidities and outcome associated with a-ANCA. Materials and methods: This retrospective study enrolled 164 and 170 patients with typical ANCA and a-ANCA positivity, respectively, who visited Taichung Veterans General Hospital, Taiwan from January 2016 to March 2021. Logistic regression analysis was used to determine risk factors and the rheumatological diagnosis associated with a-ANCA. Cox proportional hazard regression and Kaplan-Meier curves were employed to identify variables associated with 5-year renal survival and mortality. Results: Patients with a-ANCA had lower chance of ANCA-associated vasculitis (OR: 0.02, 95 % CI: 0.01-0.07 p < 0.001), and systemic lupus erythematosus (OR: 0.23, 95 % CI: 0.11-0.48, p < 0.001), but a higher risk of rheumatoid arthritis (OR: 2.99, 95 % CI: 1.15-7.83, p = 0.025) and ulcerative colitis (OR: 5.50, 95 % CI: 1.20-25.29, p = 0.028). Patients with a-ANCA had a better renal survival (OR: 0.14, 95 % CI: 0.08-0.24, p < 0.001) and lower mortality (OR: 0.31, 95 % CI: 0.16-0.60, p = 0.001) than patents in the typical ANCA group. The 5-year renal survival and mortality was 89.3 % and 8.8 %, respectively, in patients with a-ANCA. Conclusion: Patients with a-ANCA had better renal survival and lower mortality rates compared to patients with typical ANCA. These real-world data provide evidence of the long-term outcome and shed light on avenues for the strategic management of patients with a-ANCA.

Application of Supervised Machine Learning to Recognize Competent Level and Mixed Antinuclear Antibody Patterns Based on ICAP International Consensus.

BACKGROUND: Antinuclear antibody pattern recognition is vital for autoimmune disease diagnosis but labor-intensive for manual interpretation. To develop an automated pattern recognition system, we established machine learning models based on the International Consensus on Antinuclear Antibody Patterns (ICAP) at a competent level, mixed patterns recognition, and evaluated their consistency with human reading. METHODS: 51,694 human epithelial cells (HEp-2) cell images with patterns assigned by experienced medical technologists collected in a medical center were used to train six machine learning algorithms and were compared by their performance. Next, we choose the best performing model to test the consistency with five experienced readers and two beginners. RESULTS: The mean F1 score in each classification of the best performing model was 0.86 evaluated by Testing Data 1. For the inter-observer agreement test on Testing Data 2, the average agreement was 0.849 (κ) among five experienced readers, 0.844 between the best performing model and experienced readers, 0.528 between experienced readers and beginners. The results indicate that the proposed model outperformed beginners and achieved an excellent agreement with experienced readers. CONCLUSIONS: This study demonstrated that the developed model could reach an excellent agreement with experienced human readers using machine learning methods.