Synopsis

Ultra High Field MRI gives the opportunity to study in vivo articular cartilage with optimal signal-to-noise ratio, contrast-to-noise ratio and spatial resolution in comparison with lower field strength systems. We obtained the mean T2 and T2* maps values for under and over 50 years aged healthy subjects, both mean values increase with the subjects' age. With dedicated multichannel coils and specific cartilage sequences, it is possible to make both a qualitative and a quantitative evaluation of early and pre-clinical cartilage ageing-related changes in estimating the water content and the integrity of the cartilage matrix.

Introduction

MRI has been established as the standard cartilage imaging technique and it is the only modality that allows direct visualization of cartilage with sufficient contrast. It has already been recognized as the most important non-invasive diagnostic modality in particular for the assessment of anatomical and morphological cartilage status. Anyway changes in cartilage physiology prior to morphological ones, cannot be visualized or measured with conventional MRI, although detecting these subtle changes is a priority due to the importance of diagnosing and managing OA at its earliest stages¹. For their suitability for quantitative assessment of cartilage status, MR sequences have to present high SNR, CNR and high-spatial resolution; they should minimize image distortions and should be fast. Using 7 T MRI permits to reach the magnification of some characteristics necessarily required for cartilage imaging. The larger susceptibility effect of UHF results in more informative images with an increased contrast between different components of the same volume; the linear SNR increase with the field strength guarantees a better spatial resolution and a faster scanning speed².

Purpose

Materials and methods

We have recruited 17 healthy subjects of different ages (<50 yrs, n=10, mean age: 39; >50 yrs, n=7, mean age= 57). Inclusion criteria were absence of clinical symptoms and no history of previous knee surgery or significant knee trauma. All the subjects underwent MR knee examination at 7 T whole-body system. We used a 8-channel volume coil, called Kneena, created by IMAGO 7 team, to explore the whole knee joint. MR protocol included: 3D bSSFP fast imaging employing steady-state acquisition (FIESTA), multiple-echo recombined gradient echo (MERGE), coherent oscillatory state acquisition for the manipulation of image contrast (COSMIC), 3D T2 star-weighted angiography (SWAN) for T2* Mapping, multi-echo SE T2 for T2 Mapping, Dixon-type pulse sequence Iterative Decomposition of water and fat with Echo Asymmetry and Least-squares estimation (IDEAL) for fat-saturation images. For T2 relaxation, a sagittal, me-SE sequence with 8 echoes was performed, whereas T2* relaxation was obtained by an axial 3D T2 SWAN. For each subject, T2 and T2* mapping were obtained with an homemade software in Matlab, respectively in sagittal and axial projection, including patellar, femoral and tibial cartilage. We have manually drawn the region of interest (ROI) on the maps, at the maximum thickness of the cartilage, using Horos³ program. In the T2* maps for the layout of the tibial articular cartilage, we have drawn the ROIs only on the patellar and femoral cartilage excluding the tibial compartment, which generates errors because its orientation is parallel to the slices acquired. Statistical data analysis was performed using JMP version 7.0 (SAS) statistics software. One-way Anova analysis of variance was used to compare the different distribution of T2 and T2* map values between groups. A statistical comparison between age and measurements was performed by bivariate test, and a p-value of 0.05 or less was considered to indicate significance.

Results

Conclusions

Acknowledgements

References

1. Link TM, Stahl R, Woertler K. Cartilage imaging: motivation, techniques, current and future significance Eur Radiol (2007) 17: 1135–1146

2. Moser E. et al. Ultra-high-field magnetic resonance: Why and when? World J Radiol 2010; 2(1): 37-40

3. Horos Project (2015). DICOM image viewing and measuring. [Horos]. http://www.horosproject.org/