To evaluate the ability of depiction of deep fascia, epimysium and periosteum in 3 dimensional Ultrashort Echo Time MR Imaging (3D UTE) with histologic correlation in porcine model.

This experimental study was exempt from the institutional review board approval and informed consent was not required. Three fresh porcine lower legs (Two were right leg and on were left leg) were obtained within 24 hours of death. They were dissected and prepared as follows (Fig 1): zone 1: superficial soft tissue from skin to subcutaneous layer was flapped and soft tissue from deep fascia remained, zone 2: superficial soft tissue from skin to deep fascia was flapped and soft tissue from epimysium remained, zone 3: superficial soft tissue from skin to epimysium was flapped and soft tissue from muscle was remained, zone 4: superficial soft tissue from skin to superficial portion of muscle was flapped and deep portion of muscle remained, zone 5: soft tissue from skin to periosteum was stripped from medial cortex of tibia. Aqaurius gel put between these flaps and remaining tissues. One of porcine tibia was dissected and extracted from soft tissue. The tibia was divided into 3 segments as follows: 1) bone with intact periosteum 2) bone with stripping of periosteum, 3) bone with stripping of periosteum and scratching outer surface of cortex. It was placed into a container filled with gelatin. Before and after preparing all samples, a basic 3D UTE sequence was implemented on a 3-T MR scanner (Discovery 750 W, GE Healthcare, Waukesha, WI) with a maximum gradient performance of 44 mT/m and 200 mT/m/ms using a medium-sized phase-array coil (Gem Flex Suite). For the contrast for short T2 tissue, the double echo acquisition with later echo subtraction technique was used to suppress long T2 signals. The first FID acquisition with 28 μs detects signal from both long and short T2 components, while the second echo with 4.4 ms detects signal from longer T2 components. Subtraction of the second echo from the first one selectively suppresses signal from the longer T2 components and typically provides high contrast in the short T2 range1. A histologist evaluated the histologic slides of all flaps and the extracted porcine tibia for presence or absence of deep facia, epimysium and periosteum. Two musculoskeletal radiologists reviewed with consensus the 3D UTE images with subtraction.The deep fascia was clearly seen as a high signal intensity at the side of remaining tissue in zone 1, and at the side of flap in zone 2 in 3D UTE images with subtraction and confirmed by histologic slide (Fig 2). The epimysium was seen as a subtle high signal intensity at the remaining tissue of zone 2 (Fig 2) in both 3D UTE images with subtraction and histologic slide. However, in zone 3, epimysium was not seen in either flap and remaining tissue in 3D UTE images with subtraction in contrast to histologic slide (Fig 3). The surface of muscle showed no high signal intensity in either flap or remaining tissue (Fig 3). The stripped periosteum was seen as a high signal intensity and confirmed by histologic slide (Fig 3). However, high signal intensity was also seen at the remaining bone (Fig 3). The extracted tibia showed thick high signal intensity around cortex at the segment which had intact periosteoum including both outer fibrous and inner osteogenic layer (Fig 4). Bone with stripping of periosteum showed reduced but remaining high signal intensity around the cortex. Corresponding histologic slide revealed remaining inner osteogenic layer of periosteum (Fig 4). The bone with stripping periosteum and scratching outer surface of cortex also showed reduced but remaining high signal intensity around cortex. Corresponding histologic slide confirmed complete naked cortex (Fig 4).The 3D UTE might depict deep fascia and periosteum as high signal intensities whereas the ability of visualization of epimysium was uncertain and high signal intensity of artifacts remain in naked cortex without periosteum.In our study, 3D UTE successfully demonstrated the deep fascia and periosteum. However, the epimysium was not showed clearly because it was too thin to be visualized. The main source of high signal intensity of periosteum in 3D UTE might arise from the outer fibrous layer. The same feature, reduced but remaining high signal intensity around cortex, in cortex with ostegenic layer and in naked cortex indicate that 3D UTE cannot detect the signal from osteogenic layer and some kinds of artifacts might exist around the cortical bone. We assumed these artifacts to be susceptibility artifacts or out of slice effect 2.