Evaluation of femoral perfusion in a non-traumatic rabbit osteonecrosis model with T2*-weighted dynamic MRI.

We evaluated femoral perfusion in a non-traumatic rabbit serum sickness osteonecrosis (ON) model, using serial repetitive T2*-weighted (T2*W) dynamic magnetic resonance imaging (MRI) and investigated prediction of ON occurrence in early stages, comparing T2*W dynamic MRI with non-enhanced (T2-, T1- and fat suppression T1-weighted) and contrast-enhanced MRI. Early microcirculatory injury or necrotic lesion was detected in 0% of femora (extravasation, 0/6) at 72 h, 33% (necrotic, 4/12) at 1 week and 100% (necrotic, 14/14) at 3 weeks using non-enhanced MRI, and in 67% of femora (extravasation, 4/6) at 72 h, 58% (necrotic, 7/12) at 1 week and 100% (necrotic, 14/14) at 3 weeks using contrast-enhanced MRI. In contrast, microcirculatory injury or necrotic lesion was detected in 83% of femora (extravasation, 5/6) at 72 h, 92% (necrotic, 11/12) at 1 week and 100% (necrotic, 14/14) at 3 weeks using T2*W dynamic MRI as no transient decrease or less marked transient decrease in signal intensity of regions of interest (ROIs), compared to normal femora, which showed a clear transient decrease in signal intensity of ROIs. These results indicate that T2*W dynamic MRI with optimal imaging parameters and a dose of contrast agent is the most sensitive of these three MRI methods and may be clinically useful for evaluating femoral perfusion in artery phase and predicting ON occurrence.