Development of a scoring method to visually score cortical interruptions on high-resolution peripheral quantitative computed tomography in rheumatoid arthritis and healthy controls.

Abstract

To develop a scoring method to visually score cortical interruptions in finger joints on High-Resolution peripheral Quantitative Computed Tomography (HR-pQCT), determine its intra- and inter-reader reliability and test its feasibility. The scoring method was developed by integrating results from in-depth discussions with experts, consensus meetings, multiple reading experiments and the literature. Cortical interruptions were scored by two independent readers in an imaging dataset with finger joints from patients with rheumatoid arthritis (RA) and healthy controls and assessed for adjacent trabecular distortion. Reliability for the total number of cortical interruptions per joint and per quadrant was calculated using intraclass correlation coefficient (ICC). Feasibility was tested by recording the time to analyze one joint. In 98 joints we identified 252 cortical interruptions, 17% had trabecular distortion. Mean diameter of the interruptions was significantly larger in patients with RA compared with healthy controls (0.88 vs 0.47 mm, p = 0.03). Intra-reader reliability was ICC 0.88 (95% CI 0.83;0.92) per joint and ICC 0.69 (95% CI 0.65;0.73) per quadrant. Inter-reader reliability was ICC 0.48 (95% CI 0.20;0.67) per joint and ICC 0.56 (95% CI 0.49;0.62) per quadrant. The time to score one joint was mean 9.2 (SD 4.9) min. This scoring method allows detection of small cortical interruptions on HR-pQCT imaging of finger joints, which is promising for use in clinical studies.