Yumi Park1, Seon Young Kim1, Gye Cheol Kwon1, Sun Hoe Koo1, Eun-Suk Kang2, Jimyung Kim3. 1. Department of Laboratory Medicine, Chungnam National University Hospital, Daejeon, Republic of Korea. 2. Departments of Laboratory Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea. 3. Department of Laboratory Medicine, Chungnam National University Hospital, Daejeon, Republic of Korea jmkim@cnuh.co.kr.
Abstract
BACKGROUND: The detection of antinuclear antibodies (ANAs) is crucial for the diagnosis of systemic autoimmune rheumatic diseases. Here, we evaluated the performance of an automated ANA image analyzer, EUROPattern Suite. METHODS: We compared the automated and manually revised interpretations of EUROPattern Suite with conventional microscopic interpretations of indirect immunofluorescence (IIF) by calculating the analytical sensitivity and specificity for ANA detection and evaluating the concordance between these approaches for ANA detection, pattern recognition, and titer determination. RESULTS: The analytical sensitivity and specificity of EUROPattern Suite for ANA detection were 97.7% and 88.2%, respectively, relative to those of conventional IIF, with a concordance of 94.4% 3κ=0.874). After manual revision, the analytical sensitivity, specificity, and concordance improved to 98.5%, 94.1%, and 97.0% (κ=0.932), respectively. The concordance rates for pattern recognition with simple and mixed patterns were 72.1% and 56.0%, respectively, and varied depending on the type of pattern. After manual revision, the concordance rates with simple and mixed patterns increased to 97.1% and 84.0%, respectively. The concordance rate for titer determination with a simple pattern was 84.6%. CONCLUSION: Although automated pattern recognition has some limitations, EUROPattern Suite facilitates automated ANA detection with a high sensitivity and concordance with conventional IIF.
BACKGROUND: The detection of antinuclear antibodies (ANAs) is crucial for the diagnosis of systemic autoimmune rheumatic diseases. Here, we evaluated the performance of an automated ANA image analyzer, EUROPattern Suite. METHODS: We compared the automated and manually revised interpretations of EUROPattern Suite with conventional microscopic interpretations of indirect immunofluorescence (IIF) by calculating the analytical sensitivity and specificity for ANA detection and evaluating the concordance between these approaches for ANA detection, pattern recognition, and titer determination. RESULTS: The analytical sensitivity and specificity of EUROPattern Suite for ANA detection were 97.7% and 88.2%, respectively, relative to those of conventional IIF, with a concordance of 94.4% 3κ=0.874). After manual revision, the analytical sensitivity, specificity, and concordance improved to 98.5%, 94.1%, and 97.0% (κ=0.932), respectively. The concordance rates for pattern recognition with simple and mixed patterns were 72.1% and 56.0%, respectively, and varied depending on the type of pattern. After manual revision, the concordance rates with simple and mixed patterns increased to 97.1% and 84.0%, respectively. The concordance rate for titer determination with a simple pattern was 84.6%. CONCLUSION: Although automated pattern recognition has some limitations, EUROPattern Suite facilitates automated ANA detection with a high sensitivity and concordance with conventional IIF.