Magnetic resonance imaging evaluation after implantation of a titanium cervical disc prosthesis: a comparison of 1.5 and 3 Tesla magnet strength

Abstract

Purpose: Cervical disc prostheses induce significant amount of artifact in magnetic resonance imaging which may complicate radiologic follow-up after surgery. The purpose of this study was to investigate as to what extent the artifact, induced by the frequently used Discover(®) cervical disc prosthesis, impedes interpretation of the MR images at operated and adjacent levels in 1.5 and 3 Tesla MR.

Methods: Ten subsequent patients were investigated in both 1.5 and 3 Tesla MR with standard image sequences one year following anterior cervical discectomy with arthroplasty.

Outcome measures: Two neuroradiologists evaluated the images by consensus. Emphasis was made on signal changes in medulla at all levels and visualization of root canals at operated and adjacent levels. A "blur artifact ratio" was calculated and defined as the height of the artifact on T1 sagittal images related to the operated level.

Results: The artifacts induced in 1.5 and 3 Tesla MR were of entirely different character and evaluation of the spinal cord at operated level was impossible in both magnets. Artifacts also made the root canals difficult to assess at operated level and more pronounced in the 3 Tesla MR. At the adjacent levels however, the spinal cord and root canals were completely visualized in all patients. The "blur artifact" induced at operated level was also more pronounced in the 3 Tesla MR.

Conclusions: The artifact induced by the Discover(®) titanium disc prosthesis in both 1.5 and 3 Tesla MR, makes interpretation of the spinal cord impossible and visualization of the root canals difficult at operated level. Adjusting the MR sequences to produce the least amount of artifact is important.

Fig. 1

a Left 1.5 T MRI with T2 sagittal images. Right 1.5 T MRI with T2 sagittal images and with an artists illustration of the artifact around the disc prosthesis and effect on the spinal cord (illustrated in yellow). The artifact gives the impression that the spinal cord is pulled in the anterior direction towards the disc space/prosthesis and with a change in configuration. b Left 3 T with T2 sagittal images. Right 3 T with T2 sagittal images and with an artists illustration of the artifact around the disc prosthesis and effect on the spinal cord (illustrated in yellow) The artifact gives the impression that the spinal cord is dislocated in the posterior direction and with a signal change within the cord. The artifact can be misinterpreted as spinal stenosis

Fig. 2

Left a and b T1 sagittal images (a 1.5 T and b 3 T) with the hight of the artifact and the hight between the upper and lower endplate for the two adjacent vertebrae shown with lines. Right a and b the artists illustration of the amount of artifact produced around the titanium prostheses in a 1.5 and b 3 T MRI. The artifact ratios were calculated by measuring the hight of the artifact cranial and caudal to the prostheses measured at the center of the vertebrae divided with the hight between the upper and lower endplate for the two adjacent vertebrae

Fig. 3

a 1.5 T oblique view showing the artifact at operated level C5/C6 and adjacent upper and lower-level root canals. b 3 T oblique view showing the artifact at operated level C5/C6 and adjacent upper and lower-level root canals