Visual detection of cortical breaks in hand joints: reliability and validity of high-resolution peripheral quantitative CT compared to microCT

BACKGROUND To study the reliability and validity of high-resolution peripheral quantitative CT (HR-pQCT) with microCT (μCT) as gold standard in the visual detection of cortical breaks in metacarpophalangeal (MCP) and proximal interphalangeal (PIP) joints. METHODS Ten cadaveric fingers (10 MCP and 9 PIP joints) were imaged by HR-pQCT and μCT and visually analyzed by two independent readers. Intra- and interreader reliability were evaluated for the presence (yes/no, kappa statistics) and the total number (intraclass correlation coefficient, ICC) of cortical breaks. Sensitivity, specificity, positive and negative predictive value (PPV respectively NPV) of HR-pQCT in detecting cortical breaks were calculated. RESULTS With HR-pQCT, mean 149 cortical breaks were identified and with μCT mean 129 (p < 0.05). Intrareader reliability for the presence of a cortical break per quadrant was 0.52 (95 % CI 0.48-0.56) and 0.71 (95 % CI 0.67-0.75) for HR-pQCT and μCT, respectively, and for the total number of cortical breaks 0.61 (95 % CI 0.49-0.70) and 0.75 (95 % CI 0.68-0.82). Interreader reliability for the presence of a cortical break per quadrant was 0.37 (95 % CI 0.33-0.41) and 0.45 (95 % CI 0.41-0.49) for HR-pQCT and μCT, respectively, and for the number of cortical breaks 0.55 (95 % CI 0.43-0.65) and 0.54 (95 % CI 0.35-0.67). Sensitivity, specificity, PPV and NPV of HR-pQCT were 81.6, 64.0, 81.6, and 64 % respectively. CONCLUSION Cortical breaks were commonly visualized in MCP and PIP joints with HR-pQCT and μCT. Reliability of both HR-pQCT and μCT was fair to moderate. HR-pQCT was highly sensitive to detect cortical breaks with μCT as gold standard.