Dependence of a Load-induced Laminar Appearance in Articular Cartilage on MRI Echo Time

Articular cartilage is mainly composed of water, collagen fibers, and glycosaminoglycans. In several previous studies, an unusual laminar appearance was discovered in the deep region of articular cartilage when the tissue was oriented at the magic angle and compressed, which revealed the load-induced deformation of the collagen matrix (1, 2).

Aims: To study the orientation-dependent properties of cartilage using Ultra-short echo time (UTE) and Multiple-slice-multiple-echo (MSME) sequences, under external loading and in two soaking solutions (normal saline, phosphate-buffered saline (PBS)).

Materials and Methods: Cartilage specimens were loaded at different strains, soaked in saline and PBS, and oriented at various angles respect to B0 (0°, 30°, 55°, 90°, 125°, 150°, 180°, 210°, 235°, 270°, 305°, 330°, 360°). µMRI used a Bruker micro-imager (7T), with the UTE and MSME sequences. The echo times of UTE and MSME were 200 µs and 3.0 ms, respectively.

Results: In MSME images, cartilage in saline appeared homogeneous at the magic angles (55°, 125°), but laminar at other orientations. In UTE images, cartilage in saline did not appear laminar regardless of the orientation. The load-induced laminar layer was clearly visible in MSME images, but largely reduced in UTE images. The immersion of the specimens in PBS resulted in a relatively homogeneous appearance in both MSME and UTE images. Quantitative measurements of the total cartilage thickness with and without loading depended on imaging sequences and specimen orientations. In saline with MSME, cartilage thickness at different orientations varied by nearly 20%, which means 1/5 of the tissue was not measured in routine imaging. An even larger portion of cartilage could be missed in clinical MRI due to its longer TE. Consistent thickness determination can be obtained at all orientations with the use of UTE and PBS immersion.

Conclusions: Since cartilage has a strong T2 anisotropy that is depth-dependent, MRI using routine sequences can miss the deepest tissue that interfaces with the underlining bone. This preliminary study of cartilage using both MSME and UTE sequences reveals that as much as 20% of cartilage can be missed from these routine imaging experiments. Additional UTE experiments are currently being carried out in our lab to investigate the cartilage-bone interface, which will lead to the study of a different type of osteoarthritic triggering event, trauma.

Acknowledgements: The authors thank Drs Dieter Gross and Thomas Oerther (Bruker Germany) for their help on the UTE experiments, and NIH for an R01 grant (AR052353). NW is currently at Duke University School of Medicine (Brain Imaging and Analysis Center).

References: (1) Osteoarthritis Cartilage 2004, 12: 887. (2) JMRI 2015, in press.