Biochemical MR Imaging of the intervertebral disc by 3T-MRT

Low back pain is an extremely common problem, affecting almost three quarters of the population at some time in their life. Disc anatomy is felt to play a pivotal role in the underlying pain, yet abnormal spine and disc morphology, including disc herniation has been described in the asymptomatic population, too. Scientific studies on biochemistry and biomechanics of the disc have offered insights into structure-function-failure relationships. MR imaging is a reliable technique for morphological evaluation of spine and disc. In addition to high resolution morphologic visualization of a disc and vertebral body tissue, analysis of the ultrastructural disc matrix is also necessary as it may provide more basic information on the development of disc pathology in vivo at an early stage and may reflect mechanical load-bearing capacity of intervertebral disc. Contrast-enhanced techniques as well as sodium MR imaging have been proven to be useful in in vitro and in vivo examinations to selectively visualize proteoglycans, which represent one of the basic components of the disc matrix. The basis of these techniques was developed in our working group within two Jubilee projects (funded by Austrian National Bank) and in one ongoing FWF project approved for cartilage research. Due to the similarities between cartilage and disc tissue this knowledge will be now transferred to the intervertebral disc using a state of the art MR scanner operating at high field (3 Tesla) and cutting coil technology. The structure of the collagen fibres seems to be selectively monitored by T2-Mapping and diffusion-weighted MR imaging. As we have shown in one Jubilee project, the measurement of T1rho relaxation times in surgically untreated cartilage will bring additional information compared to T2 mapping. In the near future, the described biochemical selective MR methods may be used in routine practice for examining patients following autologous disc chondrocyte implantation (ADCT) in order to monitor the regeneration of the disc over time and to detect early complication which require intervention. Since biopsies after ACDT are considered unethical, biochemical MR imaging could provide the only non-invasive tool in the follow-up. Moreover, it could be useful in evaluating newly developed disc transplantation techniques and in optimizing rehabilitation programmes in general and also individually for each subject.

For the biochemical MR imaging of the intervertebral disc (IVD) the following MR mthods and techniques were optimized within this project: T2- Mapping, T2*- Mapping, diffusion-weighted imaging at 3 Tesla and recently sodium imaging at 7 Tesla and Chemical Exchange Saturation Transfer (CEST) at 3 Tesla In a first trial on 34 patients with lower back pain but without radicular symptoms at 3 Tesla we compared standard morphological MRI with T2 mapping. For T2 mapping all disci of the lumbar spine analysed by separation into 5 equally sized Regions Of Interest (ROI) (20% each) in the sagittal plane where the anterior and posterior 20% were defined anatomically as the Annulus fibrosus (AF) and in between the 60% represented the Nucleus pulposus (NP) of the IVD. We found a good correlation between the morphological Pfirrmann classification and and T2 relaxation time values. To further increase the sensitivity of the analysis the posterior annulus fibrosus was further subdivided into two 10% ROIs to consider the axial load on the posertior column of the spine. This analysis provided a possible marker for the early stage of disc degeneration. Furthermore we performed a comparison of quantitative T2 values of the IVD with conventional MR findings of the spine at 3 Tesla. In a total of 265 discs 39 focal herniations, 10 annular tears, 123 disc bulgings and 103 normal disci were found. We found a statistically significant difference between the nucleus pulposus T2 values in discs with annular tears and all other groups with the discs with annular tears showing significant lower T2 values compared to discs without annular tears. The difference in NP T2 values between discs with focal herniations and normal discs was also significant. However no significant difference was found between NP T2 values between disc bulging and disc herniation. Thus for the first time quantitative T2 mapping of the NP demonstrated significant differences in T2 values between different dislocations of the discs of the lumbar spine at 3 Tesla. In addition methodologically T2 mapping of the IVD was compared to T2* mapping in 30 patients with low back pain at 3 Tesla. The highest variation with an increase of T2 and T2* values from the AF to NP was found in Pfirrmann grade I and decreased with higher Pfirrmann grades II-IV. In addition to T2 mapping T2* mapping revealed further changes in the posterior annulus fibrosus. The correlation between T2 and T2* mapping showed a moderate correlation of 0.21 to 0.356. The clear differentiation between different stages of disc degeneration and the possibility of quantification by T2 and T2* mapping may provide a new tool for a more sensitive monitoring of different therapies of patients with low back pain. Finally with the world wide first sodium spine coil, implemented on our 7 Tesla MR scanner, the first sodium images of the IVD could be achieved and the required sequences could be optimized by us. In a preliminary study we compared sodium imaging of the IVD of the lumbar spine with T2 mapping and morphological grading by the Pfirmman classification. No correlation was found between sodium imaging and T2 mapping which proves that both methods detect different components of the IVD, namely T2 mapping water content and collagen fiber content and organization and sodium imaging glycosaminoglycan content.