Reproducibility of imaging human knee cartilage by delayed gadolinium-enhanced MRI of cartilage (dGEMRIC) at 1.5 Tesla

Abstract

The purpose of this study was to investigate the day-to-day reproducibility of the delayed gadolinium-enhanced MRI of cartilage (dGEMRIC) measurement at different knee joint surfaces in healthy subjects at 1.5 Tesla (T). The dGEMRIC experiment was repeated for 10 asymptomatic volunteers three times with an average interval of 5 days between scans. The measurement was performed from a single sagittal slice through the center of the lateral femoral condyle and from the center of the patella in the axial plane. Cartilage was manually segmented into superficial, deep and full-thickness regions of interests (ROIs) at different topographical locations of the femur, tibia and patella. The reproducibility was evaluated separately for each ROI as well as for the entire bulk cartilage in the slice of each joint surface. The reproducibility at various ROIs expressed by root-mean-square average coefficient of variation (CV(RMS)) ranged between 4.7-12.9%. Thirty out of thirty-three ROIs showed a CV(RMS) less than 10%. Intraclass correlation coefficient (ICC) ranged between 0.45 and 0.98. The CV(RMS) and ICC for bulk dGEMRIC were 4.2% and 0.95 for femur, 5.5% and 0.87 for tibia, and 4.8% and 0.97 for patella. The dGEMRIC technique showed good day-to-day reproducibility, on the average 8% for small deep or superficial segments, 7% for full-thickness ROIs and 5% for bulk ROIs covering all visible cartilage in a single joint surface. We conclude that dGEMRIC imaging at field strength 1.5 T can be used as a reliable instrument for the assessment of articular cartilage when staff has been carefully trained.