5185

Single- and Bi-component Analyses of T2* Relaxation in Knee Tendon and Ligament by Using 3D Ultrashort Echo Time Cones (UTE Cones) Magnetic Resonance Imaging
Yinghua Zhao1, Yajun Ma2, Yanchun Zhu2, Shaolin Li1, and Jiang Du2

1Third Affiliated Hospital of Southern Medical University (Academy of Orthopedics · Guangdong Province), Guangzhou, China, 2University of California, San Diego, San Diego, CA, United States

Synopsis

For patellar tendon (PT), anterior cruciate ligament (ACL) and posterior cruciate ligament (PCL) of the knee joint, the in vivo application of the bound and free water mapping techniques is still limited. In this study, we aimed to develop 3D multi-echo fat saturated ultrashort echo time Cones (3D FS-UTE-Cones) imaging protocol for fast volumetric mapping of free and bound water components of whole knee joints in a clinical 3T scanner. The results showed that, for PT, PCL and ACL, the short and long T2* components and their fractions can be characterized by 3D FS-UTE-Cones acquisitions with bi-component T2* analysis, accompanying with superior resolution and short scanning time.

INTRODUCTION

Most knee joint tissues, including PT, PCL and ACL, have two components, namely bound water (BW) and free water (FW). FW has a longer T2* (T2*L) and is located between the network of interwoven collagen bundles, and BW has a shorter T2* (T2*S) and is associated with collagen and/or proteoglycan. 1-4 Previous studies have demonstrated that, using UTE techniques, single- and bi-exponential T2* signal decay models are suitable for studying the composition of normal and abnormal tendons and ligaments (e.g., ligament tears). 1,5 However, bi-component analysis typically requires a long scan time to allow acquisition of all images at different TEs. 6,7 High spatial resolution is also needed in order to image the fine structures in the knee joint. As a result, the in vivo application of BW and FW mapping techniques is still limited. 8-10 In this study, we aimed to develop 3D multi-echo fat saturated ultrashort echo time Cones (3D FS-UTE-Cones) imaging protocol for fast volumetric mapping of water components of the knee joint in a clinical 3T scanner.

METHODS

Knees of five healthy volunteers and six knee joint samples from cadavers were scanned with a 16-channel knee coil via 3D multi-echo FS-UTE-Cones acquisitions (Figure 1) on a clinical scanner. Single-component fitting of T2*M, and bi-component fitting of T2*S, long T2*L, short T2* fraction (Frac_S) and long T2* fraction (Frac_L), were performed within tendons and ligaments. An independent sample t-test with equal variances was performed to obtain the difference in the root mean square error (RMSE), T2*M, T2*S and T2*L values, and their fractions between PT, ACL and PCL. P less than 0.05 was considered statistically significant. Samples were stained with hematoxylin and eosin (H&E) for histology.

RESULTS

Our results showed that RMSE values of bi-exponential fits were lower than ones of single-exponential fits for all tendons and ligaments (P = 0.01), except for ACL in volunteer knee joints (P = 0.29) (Table 1). Two cases, for example, were shown in Figure 2 and 3. For knee joint samples, there was no statistical difference among all data in PT, PCL and ACL (all P > 0.05), which have been proved to be normal tissues using histology. However, for volunteers, all parameters of bi-component fitting were statistically different across PT, PCL, and ACL, except for T2*S and T2*L between PT and PCL (P = 0.63; P = 0.47, respectively), and between PCL and ACL (P = 0.88; P = 0.41, respectively). T2*S, T2*L, Frac_L, and T2*M were smaller in PT and PCL than in ACL. On the contrary, Frac_S was higher in PT and PCL than in ACL (90.54 ± 2.69% in PT, 87.02 ± 3.85% in PCL, and 21.91 ± 9.05% in ACL) (Table 2).

DISCUSSION AND CONCLUSIONS

We achieved bi-exponential T2* fitting performance with very small RMSE values in all measurements, except for that in ACL of volunteers. These results might be explained by the fact that ACL is oriented closer to the magic angle (~54°) relative to the B0 field, and thus exhibits greatly resulted in flawed measurements of ACL. In addition, the relationship between magnetization fractions and matrix components may be affected by proton exchange between compartments. 11,12 Our results suggest that multi-echo 3D UTE Cones acquisitions have some advantages over existing technologies. 3D UTE cones sequences are much less prone to eddy current artifact compared with 2D UTE sequences with half-pulse excitation, where mapping of BW and FW components may suffer from errors due to out-of-slice signal contaminations. 13 In addition, 3D Cones UTE sequences have more desirable properties for UTE MRI than 3D projection reconstruction (3DPR), such as higher SNR efficiency, less aliasing artifacts and reduced scan time. 14,15 Finally, the multi-echo acquisitions in these sequences allow relatively short scan times of 18 minutes for volumetric coverage and high resolutions mapping of T2*s and relative fractions. However, Our study has several limitations. First, 3D UTE requires longer scan times. The increased likelihood of patient movement increases susceptibility to motion artifacts, and could introduce errors in bi-exponential T2* mapping. Movement of the subjects was minimized by careful knee fixation, and the images were co-registered in post-processing. Second, the number of volunteers was small, consisting of five knees from five healthy volunteers. With more volunteers, we expect that the RMSE of ACL would be decreased, that the difference between ALC and PCL would reach significance, and that clinical diagnostic guidelines for making decisions about disorders of PT, PCL and ACL would be found using 3D UTE Cones MR acquisition. For clinical use, further optimization of the imaging protocol would be needed, including reduction of the total scan time for clinical use via parallel imaging and compressed sensing techniques.

Acknowledgements

The authors thanked Niloofar Shojaeiadib for the statistical analysis, and Rose Luo and Jonathan Wong for proofreading the manuscript. This study has received grants from the National Scientific Foundation of China (No. 81471810), the Science and Technology Plan Projects of Guangdong provincial, China (No. 2014A020211018, 2014A020212399).

References

1. Kijowski R, Wilson JJ, Liu F. Bicomponent ultrashort echo time T2* analysis for assessment of patients with patellar tendinopathy. Journal of Magnetic Resonance Imaging. 2017.

2. Juras V, Apprich S, Szomolanyi P, Bieri O, Deligianni X, Trattnig S. Bi-exponential T2* analysis of healthy and diseased Achilles tendons: an in vivo preliminary magnetic resonance study and correlation with clinical score. European radiology. 2013;23(10):2814-22.

3. Robson M, Benjamin M, Gishen P, Bydder G. Magnetic resonance imaging of the Achilles tendon using ultrashort TE (UTE) pulse sequences. Clinical radiology. 2004;59(8):727-35.

4. Eckstein F, Cicuttini F, Raynauld J-P, Waterton J, Peterfy C. Magnetic resonance imaging (MRI) of articular cartilage in knee osteoarthritis (OA): morphological assessment. Osteoarthritis and cartilage. 2006;14:46-75.

5. Rahmer J, Börnert P, Dries SP. Assessment of anterior cruciate ligament reconstruction using 3D ultrashort echo‐time MR imaging. Journal of Magnetic Resonance Imaging. 2009;29(2):443-8.

6. Du J, Diaz E, Carl M, Bae W, Chung CB, Bydder GM. Ultrashort echo time imaging with bicomponent analysis. Magnetic resonance in medicine. 2012;67(3):645-9.

7. Juras V, Apprich S, Zbýň Š, Zak L, Deligianni X, Szomolanyi P, et al. Quantitative MRI analysis of menisci using biexponential T2* fitting with a variable echo time sequence. Magnetic resonance in medicine. 2014;71(3):1015-23.

8. Bae WC, Du J, Bydder GM, Chung CB. Conventional and ultrashort time-to-echo magnetic resonance imaging of articular cartilage, meniscus, and intervertebral disk. Topics in magnetic resonance imaging : TMRI. 2010;21(5):275-89.

9. Pauli C, Bae WC, Lee M, Lotz M, Bydder GM, D'Lima DL, et al. Ultrashort-echo time MR imaging of the patella with bicomponent analysis: correlation with histopathologic and polarized light microscopic findings. Radiology. 2012;264(2):484-93.

10. Chang EY, Du J, Iwasaki K, Biswas R, Statum S, He Q, et al. Single- and Bi-component T2* analysis of tendon before and during tensile loading, using UTE sequences. J Magn Reson Imaging. 2015;42(1):114-20.

11. Bydder M, Rahal A, Fullerton GD, Bydder GM. The magic angle effect: a source of artifact, determinant of image contrast, and technique for imaging. Journal of magnetic resonance imaging : JMRI. 2007;25(2):290-300.

12. Robson MD, Gatehouse PD, Bydder M, Bydder GM. Magnetic resonance: an introduction to ultrashort TE (UTE) imaging. Journal of computer assisted tomography. 2003;27(6):825-46.

13. Josan S, Kaye E, Pauly JM, Daniel BL, Pauly KB. Improved half RF slice selectivity in the presence of eddy currents with out-of-slice saturation. Magn Reson Med. 2009;61(5):1090-5.

14. Gurney PT, Hargreaves BA, Nishimura DG. Design and analysis of a practical 3D cones trajectory. Magnetic resonance in medicine. 2006;55(3):575-82.

15. Barger AV, Block WF, Toropov Y, Grist TM, Mistretta CA. Time-resolved contrast-enhanced imaging with isotropic resolution and broad coverage using an undersampled 3D projection trajectory. Magn Reson Med. 2002;48(2):297-305.

Figures

Figure 1. The 3D UTE Cones sequence (A). After excitation with a short rectangular pulse, a 3D Cones trajectory (B) is used to allow time-efficient sampling with a minimal TE of 32 µs.

Figure 2. 3D UTE Cones images and region-of-interest (ROI) shown in a patella tendon (PT) sample with red lines, a posterior cruciate ligament (PCL) sample with yellow lines and an anterior cruciate ligament (ACL) sample with blue lines (A), following with histology in the PT(B), PCL (C) and ACL (D), where collagen is arranged in tightly cohesive well-demarcated bundles (Stain: hematoxylin and eosin; original magnification, *200), as well as single- and bi-component fitting (E, F, G) of multi-echo UTE image acquired at TE (0.2/3.3/15 ms, 0.5/5.5/20 ms, 0.8/8/25 ms, 2.1/11/30 ms of a 45 years old female cadaver.

Figure 3. Selected 3D UTE Cones images and region-of-interest (ROI) shown with red lines in patella tendon (PT) (A), posterior cruciate ligament (PCL) (C) and anterior cruciate ligament (ACL) (E), as well as single- and bi-component fitting (B, E, F) of interleaved multi-echo UTE image acquired at TE (0.032/4.4/20/40 ms, 0.4/6.6/25/50 ms, 0.8/1/30/60 ms, 2.2/16/35/70 ms of a 29 years old male volunteer. All bi-component fitting shows superior over single-component fitting. Dashed lines represent the estimated T2* curve and solid black circles represent the data points.

P<0.05; RMSE: Root mean square error; PT: patella tendon; ACL: anterior cruciate ligament; PCL: posterior cruciate ligament.

SD: standard deviation; PCL: posterior cruciate ligament; ACL: anterior cruciate ligament; PT: patellar tendon. For knee joint samples, * P > 0.05 vs. PT; # P > 0.05 vs. PCL. For volunteer knee joints, all P < 0.05, except for † P > 0.05 vs. PT, and ‡ P > 0.05 vs. PCL

Proc. Intl. Soc. Mag. Reson. Med. 26 (2018)
5185