MRI of acute bone bruises: Timing of the appearance of findings in a swine model

OBJECTIVE. We developed an animal model of bone injury to determine the timing of the MR appearance of bone bruises and to follow these changes over time. MATERIALS and METHODS. We performed serial MRI of 16 knees of eight swine with one nontraumatized knee as a control and 15 knees traumatized by direct patellar impact injuries using a force-calibrated device. All knees were imaged on a 1.5-T scanner using an 8-channel phased-array coil with T1-weighted, fat-saturated T2-weighted and STIR sequences. Scanning was performed at 1, 6, 12, and 30 hours after impact injury. Two radiologists independently reviewed each MR examination to identify MR signal intensity changes in the patellae and adjacent femoral condyles. RESULTS. In the 15 traumatized knees, bone bruises were noted in 93% of the patellae on T1-weighted images and in 87% of the patellae on fluid-sensitive MR images at 1 hour after injury and in 100% of the patellae at 6 hours. T1-weighted images were insensitive for detection of bone bruises in the femoral condyles. Bone bruises in the femoral condyles were seen on fluid-sensitive MR sequences as early as 1 hour after injury, with an increasing frequency over the 30-hour period. CONCLUSION. Bone bruises can be seen as soon as 1 hour after trauma but may not be seen until 30 hours after trauma. Fluid-sensitive ( fat-saturated, T2-weighted, and STIR) MR sequences are more sensitive than T1-weighted images in the detection of bone bruises.