MR imaging of the cervical spine: techniques for two- and three-dimensional imaging.

This review briefly describes two-dimensional (2D) and three-dimensional (3D) MR imaging of the cervical spine and focuses on the current and potential clinical usefulness of 3D sequences in the evaluation of cervical spine abnormalities. The practical advantages of 3D imaging are several, and include an increase in signal-to-noise ratio over 2D imaging, thin contiguous slices, more accurate slice thickness, and optimal computer postprocessing. Different 3D techniques can provide either high- or low-signal-intensity CSF, with excellent suppression of CSF pulsation artifacts.     

MR imaging of the cervical spine: neurovascular anatomy

High-resolution surface-coil MR imaging reveals intricate anatomic detail of the cervical spinal canal and its neurovascular contents. Appreciation of the normal neurovascular anatomy provides a scientific foundation for the detection of disease. Sagittal, axial, and oblique MR images of normal subjects were correlated with comparable anatomic sections obtained with a cryomicrotome whole-organ sectioning technique. The anterior epidural venous plexus is a prominent structure in the cervical spinal canal and was consistently identified both with cryomicrotomy and with MR in sagittal and axial planes. Epidural veins can be displaced and distorted in patients with cervical disk disease. Nerve roots including dorsal and ventral rootlets were consistently identified on axial images coursing through the subarachnoid space. Oblique-plane imaging showed nerve roots "en face" in their respective foramina; this may be a useful imaging technique in the diagnosis of nerve root impingement.     

MR imaging in cervical spine trauma

Continual improvements in MR imaging, technology and MR imaging-compatible monitoring and fixation devices have allowed the incorporation of this relatively new imaging modality into standard algorithms for cervical spine trauma assessment. The ability of MR imaging to define the type of spinal cord injury, the cause and severity of spinal cord compression, and the stability of the spinal column is unmatched. The heavy reliance of the spinal surgeon on MR imaging for decisions regarding the type of therapy, the timing, the approach of surgical intervention, and for predicting patient outcome attests to the usefulness of this modality.     

Oblique plane MR imaging of the cervical spine

Surface coil magnetic resonance (MR) imaging of the spine in conventional sagittal and parasagittal planes has been reported to show the cervical tissues with great clarity. Theoretically, an imaging plane perpendicular to the cervical neural foramina would be particularly effective for demonstrating the cervical spinal nerves and roots. We correlated MR images and cryomicrotome sections of the cervical spine to analyze the MR appearance of the neural foramina in this view. The normal MR appearances of the dorsal and ventral roots in cross section and, for the first time, the interradicular cleft were identified. New criteria for diagnosis of nerve root compression were suggested by the anatomic observations.     

MR imaging in acute cervical spine trauma

Fourteen patients with cervical spine injuries, 12 with resultant neurological deficits, were scanned with magnetic resonance (MR) imaging within 7 days following injury. Useful information concerning the status of the spinal canal and disks was obtained in most cases. In addition, MR was able to suggest the nature of the pathological changes within the spinal cord, as well as hemorrhage and edema in the extraspinal soft tissues. These observations indicate that following acute cervical spine trauma, MR is a valuable technique in assessing injury to the spinal cord, surrounding soft tissues, vertebra, and disks.     

MR imaging of the cervical spine in rheumatoid arthritis

The cervical spine was examined with MR imaging and conventional radiography in 23 patients with severe rheumatoid arthritis. All patients had neck pain and 17 also had neurologic symptoms. MR provided detailed information about soft-tissue lesions, vertebral dislocation, and narrowing of the spinal canal. Pannus surrounding the odontoid process was revealed in 14 patients, all with horizontal atlantoaxial subluxation. Compression of the medulla and/or spinal cord, caused by dislocated vertebrae and/or the soft-tissue mass around the odontoid process, was seen in 15 patients. When there was more than one dislocation the most important level could be determined. Posterior occipitocervical fusion had been performed in six of the patients, and in only two of these was adequate analysis of the upper cervical spine impossible because of artifacts from metal (stainless steel wires and pins). Sagittal MR in the neutral position combined with conventional radiography, including lateral views in flexion and extension, provided all the information necessary for further clinical management of rheumatoid arthritis of the cervical spine.     

Gradient-echo MR imaging of the cervical spine: evaluation of extradural disease

A prospective study was undertaken on 204 consecutive patients comparing low flip angle gradient-echo and T1-weighted spin-echo techniques in the MR evaluation of cervical extradural disease. Four patient groups were studied with varying gradient-echo TEs (6 or 13 msec) and flip angles (10 degrees or 60 degrees). Images were evaluated independently for contrast behavior and anatomy, then directly compared for conspicuity of lesions. The FLASH sequences (especially with a 10 degrees flip angle) produced better conspicuity of disease in half the imaging time. T1-weighted spin-echo sequences were more sensitive to marrow changes and intradural disease. The short TE sequence (6 msec) did not produce any diagnostic advantage over the longer TE sequence (13 msec). A fast and sensitive MR examination for cervical extradural disease combines a sagittal T1-weighted spin-echo acquisition with sagittal and axial FLASH 10 degrees sequences.     

MR imaging of intradural extramedullary tumors of the cervical spine

Magnetic resonance imaging was performed in eight patients with extramedullary cervical tumors; two were studied with a 0.15 T resistive system and six with a 0.5 superconductive unit. All tumors were verified surgically: Five meningiomas and three schwannomas were found. Magnetic resonance demonstrated the location and size of the tumors as well as their relationship with and mass effect on the adjacent spinal cord. Meningiomas showed a tendency to have relaxation times close to those of the normal parenchyma; in multiple echo sequences their signal tends to decay in a way similar to that of the spinal cord. The different signal from the surrounding cerebrospinal fluid sometimes provides a visual cleavage to define the extramedullary location of the neoplasm. Schwannomas and neurofibromas have a less constant behavior; their T2 relaxation time may be slightly longer than that of the spinal cord and the tumor may display a more intense signal in the T2 weighted images. Usually, however, the signal intensity of the extramedullary tumors is lower than that of the intramedullary.     

Asymptomatic degenerative disk disease and spondylosis of the cervical spine: MR imaging

Evidence on magnetic resonance (MR) images of disk degeneration and herniation, as well as of cord and root impingement, may be regarded either as normal, age-related changes or as causative of symptoms. Individuals referred for MR examinations of the larynx without symptoms referable to the cervical spine were studied retrospectively (35 patients) or prospectively (65 patients) over a 2-year period. With a solenoid surface coil, 5-mm-thick sections were acquired in sagittal, axial, and coronal planes with T1-weighted spin-echo pulsing sequences. Disk protrusion (herniation/bulge) was seen in five of 25 (20%) patients aged 45-54 and 24 of 42 (57%) patients older than 64 years of age. Posterolateral protrusions were seen in only nine of 100 patients and occurred with greatest frequency in patients over 64 years of age. In no patient was obliteration of the intraforaminal fat seen. Spinal cord impingement was observed in nine of 58 (16%) patients under 64 years of age, and in 11 of 42 (26%) patients over 64 years of age. Cord compression was observed in seven of 100 patients and occurred solely secondary to disk protrusion in all cases. The percentage of cord area reduction never exceeded 16% and averaged approximately 7%.     

Acute cervical spine trauma: evaluation with 1.5-T MR imaging

Twenty-one patients with acute neurologic deficits following cervical spine trauma were evaluated with magnetic resonance (MR) imaging (n = 21), computed tomography enhanced with intrathecal contrast material (CT myelography) (n = 18), myelography (n = 13), cervical spine radiography (n = 21), and intraoperative sonography (n = 7). MR imaging proved superior to other modalities in demonstrating parenchymal spinal cord injuries and cervical intervertebral disk herniation. Although both T1- and T2-weighted studies appear necessary to evaluate the anatomic relationship of the spinal cord, thecal space, intervertebral disks, and surrounding osseous and ligamentous structures, T2-weighted sequences were more sensitive than T1-weighted studies for detection of spinal cord injury. CT myelography was superior to MR imaging in demonstrating cervical spine fractures. In most cases, myelography revealed no information that was not apparent from both CT and MR imaging studies. Preliminary experience with MR imaging of acute cervical spine trauma suggests that it should be the study of choice in symptomatic patients who are otherwise clinically stable. CT may still be required in selected patients to evaluate complex fractures.     

MR imaging features of pigmented villonodular synovitis of the cervical spine

Pigmented villonodular synovitis (PVNS) is a benign proliferative disorder primarily occurring in the large joints of the appendicular skeleton such as the knee and hip joints. We present an unusual case of PVNS involving the cervical spine in an adult. MR imaging showed an enhancing mass lesion arising from the posterior elements of the cervical spine and hyperintensity on the T2-weighted images, without evidence of T2 susceptibility effects. Gross total excision of the tumor was performed, and the diagnosis was established by histopathology.     

Isthmic spondylolysis of the lumbar spine: MR imaging at 1.5 T

The appearance on magnetic resonance (MR) images of the normal pars interarticularis in 13 patients was reviewed and contrasted with that of the pars in eight patients with spondylolysis. The pars defect usually had an intermediate signal intensity with all pulse sequences; however, this intensity was somewhat variable depending on the exact ratio of cartilage, fat, and fluid within each bone defect. The pars defect was best seen with spin-echo 600/20 (repetition time msec/echo time msec) images. In three cases, out-of-phase images showed the spondylolysis best, because of extension of fat to the borders of the defect. The sagittal view allowed one to separate spondylolysis from the joint space of posterior facets since the orientation of the defects is perpendicular to the facets; thus, a common pitfall encountered with cross-sectional axial imaging techniques is avoided. MR imaging poorly delineated bone fragments around the defect, which may produce nerve root impingement, but revealed other numerous complications that occur with spondylolysis, including spondylolisthesis and herniation of the disk above.     

MR imaging of the cervical spine: assessment of image quality with parallel imaging compared to non-accelerated MR measurements

To compare the quality of cervical spine MR images obtained by parallel imaging [generalized autocalibrating partially parallel acquisition (GRAPPA)] with those of non-accelerated imaging, we conducted both phantom studies and examinations of ten volunteers at 1.5Tesla with a dedicated 12-element coil system and a head-spine-neck coil combination. Acquisitions included axial T2-weighted (T2w) images with both methods, and sagittal T2w and T1w images in vivo with the latter coil combination. Non-accelerated MRI with two averages and GRAPPA (acceleration factor 2) with two averages (GRAPPA/2AV, time reduction of approximately 50%) and four averages (GRAPPA/4AV) were compared. In the phantom, the signal-to-noise ratio of the GRAPPA/2AV was lower than that of the other two settings. In vivo, the image inhomogeneity (non-uniformity, NU) was significantly higher in T2w GRAPPA/2AV than in both other settings, and in T1w GRAPPA/2AV compared to GRAPPA/4AV. Subjectively, the delineation of anatomical structures was sufficient in all sequences. Only the spinal cord was considered to be better delineable on the non-accelerated T1w sequence compared to GRAPPA/2AV. In part, GRAPPA/4AV performed better than the other settings. The subjective image noise was lowest with GRAPPA/4AV. In cervical spine MRI, the examination time can be reduced by nearly 42% with GRAPPA, while preserving sufficient image quality.     

Postoperative cervical spine: MR assessment

The purpose of this article is to describe the variety of magnetic resonance (MR) findings in the cervical spine following anterior diskectomy and fusion (ADF), corpectomy, and postoperative complications. Toward that end, we retrospectively reviewed MR in 73 postoperative cervical spine patients and correlated the MR with their initial operative reports. Patients imaged within 1 month of ADF showed bone grafts as discrete rectangular areas of altered signal intensity within the central portions of the disk spaces. A spectrum of graft and vertebral body signal changes are seen up to 2 years following ADF. Patients studied greater than 2 years following ADF generally showed solid bony fusions without evidence of the graft or the original disk space. Bony canal stenosis at the operative site (19 of 73 cases) and disk herniations above or below the fusion sites (21 of 73 cases) were the most common postoperative findings. Hypertrophic bone was visualized as anterior extradural defect that was isointense (nine cases), hyperintense (four cases), or hypointense (eight cases) to the vertebral body on T1-weighted images. Disk herniations were evenly distributed above and below the fusion levels.     

Comparison of MR and CT myelography in imaging the cervical and thoracic spine

MR imaging and CT myelography were compared in a retrospective study of 38 patients with suspected lesions of the cervical and thoracic spinal canal and cord. Twenty-eight abnormal cases were found, including spondylosis (9), tumors (8), intramedullary cavities (3), arachnoiditis (3), disk-space-centered infection or osteomyelitis (2), nonneoplastic cord swelling (2), and CSF-borne metastasis (1). MR was equal or superior to CT myelography in depicting cases of cord enlargement, cord compression, and cord atrophy, providing better tissue characterization, no shoulder artifact, and no limitation caused by CSF block. CT myelography was superior to MR in depicting cases of spondylosis and arachnoiditis. It showed superior spatial resolution, which was most pronounced when comparing axial images and hence particularly superior in detecting the lateral extent of disk herniation. Use of surface coils and thin imaging sections is essential for accurate and complete MR evaluation of the cervical and thoracic spine.