Evaluation of a rabbit model for osteomyelitis by high field, high resolution imaging using the...

The rabbit model of osteomyelitis introduced by C.W. Norden, based on injection of an infecting solution (Staphylococcus aureus, sodium morrhuate) into the tibia, was studied at 4.7 Tesla with a time-efficient chemical shift selective imaging technique, Chemical Shift Specific Slice Selection (C4S). The evolution of the disease over several weeks was followed on water-selective, fat-selective, and sum images obtained simultaneously with this imaging sequence. Experiments were performed either on different groups of rabbits at different times after infection with subsequent sacrifice of the animal and microbiological analysis of the infected tibia or on the same group of animals imaged several times after infection. Associated analysis of the water and fat selective images revealed marrow modifications very early (Day 5 after inoculation) demonstrating the high sensitivity of the employed imaging technique. Later on, bone modifications were best identified on the sum images. Additional experiments performed on animals injected with a noninfecting solution containing only sodium morrhuate showed however that the sclerosing agent alone can yield images similar to those produced by infection at early stages after inoculation. Therefore, the Norden model would not be suitable for monitoring quantitatively outcome of therapy by magnetic resonance imaging. It is however well adapted for the evaluation and optimization of MRI techniques or protocols intended to detect early changes of bone marrow produced by septic or aseptic infarct.