Contrast-enhanced MRI of human knee cartilage Clinical applications of the novel dGEMRIC technique to study glycosaminoglycan content in articular cartilage

Sammanfattning: Delayed gadolinium-enhanced MRI of cartilage (dGEMRIC) is a new technique to study cartilage glycosaminoglycan (GAG) content. The negatively charged contrast agent Gd-DTPA2-, injected intravenously, distributes in the cartilage inversely to the likewise negatively charged GAG. The GAG concentration is reflected by the MRI parameter T1 and the Gd-DTPA2- concentration by the MRI parameter R1 (R1=1/T1). We have made T1 and R1 analyses in standardized regions of interest (ROIs) in lateral and medial femoral weight-bearing cartilage. In paper I, the distribution of Gd-DTPA2- (R1) in 19 healthy volunteers was linearly dose-related after injection of three different doses, showing that no active transportation is involved. The highest R1 was registered two hours post-contrast, suggesting this time as optimal in the clinical situation. Results indicated that the highest dose (triple dose: 0.3 mmol/kg body weight) is most sensitive to minor GAG differences and the triple dose was used in papers II-V. In paper II, 17 knees in 15 patients with normal x-ray but arthroscopically verified fibrillations in the medial or lateral femoral cartilage were investigated. R1 was higher in the diseased compartment after both 1.5 and 3 hours, with the greatest difference (31%) after 1.5 hours. In paper III, T1 was compared in 37 healthy volunteers with different levels of physical activity. Elite runners had 12% longer T1 than moderately exercising individuals, who in turn had 11% longer T1 than sedentary individuals. Results suggest that human knee cartilage adapts to exercise by increasing its GAG content. In paper IV, six investigators performed repeated ROI analyses in 12 volunteers. The intra- and inter-observer variability was low (C.V. less than 2.6%) with our standardized ROI drawing technique. In paper V, we combined GAG analysis of cartilage (dGEMRIC two hours post-contrast) and synovial fluid (biochemical) in 24 patients with an acute anterior cruciate ligament (ACL) rupture. Patients were investigated 3 weeks after the ACL injury and compared with 24 healthy controls. 22/24 patients had contusions of the lateral femoral cartilage, where T1 was 14% shorter than in controls. However, T1 was 12% shorter than controls also in the medial femoral cartilage, indicating that an ACL injury initiates a global loss of GAG from knee cartilage. The GAG concentration in the synovial fluid was increased in the ACL patients and showed a tendency to have a positive correlation with T1. In conclusion, the results of this thesis support dGEMRIC as a sensitive method to detect clinically relevant GAG differences in both healthy and diseased cartilage.