Using Dynamic MRI to Diagnose Neck Pain: The Importance of Positional Cervical Cord Compression (PC3)
It has become axiomatic that cervical pain and regional imaging often correlate poorly. Both degenerative disc disease and cervical pain are common among our patients. Yet, many have one apparently without the other. One diagnostic response is to discount the informative value of cervical imaging and rely more on history, symptoms, and physical signs. Another is to consider how to better image this area of complex anatomy, without discounting the primary importance of clinical findings.
As a dynamic structure, the cervical spine anatomy may be suspected of varying in conformation, depending on its orientation in flexion, extension, rotation, and lateral bending. However, accepted diagnostic imaging, especially by magnetic resonance imaging (MRI), most often has been static and limited to neutral positioning. Such convention ultimately may prove deceptive and fraught with bias.
Neck Pain and the Spinal Cord
General medical reviews describing “neck pain” often omit mention of any role attributable to the cervical spinal cord. Abnormalities of discs, facet joints, ligaments, nerve roots, and muscles are considered, but abnormalities of the cord itself generally are not.
Such abnormalities are discussed in reviews on cervical spondylotic myelopathy (CSM). Malcolm eloquently reviewed five different CSM syndromes: “the transverse lesion syndrome, the motor system syndrome, the central cord syndrome, the Brown Sequard syndrome, and the brachalgia and cord syndrome.”1 Each is associated with a different set of clinical signs and symptoms. Figure 1 provides an excellent example of cord compression using a traditional, static, neutral C-spine MRI protocol. In Dr. Malcolm’s review, he describes a case where the cervical spinal cord was compressed at 2 disc levels; after decompression, the cord was freed and there was evidence of cerebral spinal fluid flowing unimpeded within a widened cervical canal. If we accept these images as indicative of—or at least consistent with—CSM followed by resolution of its symptoms, then one may argue that we are compelled to similarly judge identical images obtained through dynamic imaging.
Over 50 years ago, Olsson reported how position influenced cervical spinal canal diameter in the canine model.2 Penning expanded this evaluation to human spondylitic myelopathy in 1966.3 Much later, after the discovery of MRI, Muhle et al examined a cohort of 46 patients to assess the effect of conformational changes to the cervical canal in extension (30°) and flexion (50°).4 Cervical spinal stenosis was more commonly found at extension (48%) compared to flexion (24%, P<0.05). Further, 11% of the 46 patients had cord compression in flexion compared to 20% in extension. “Concerning the number of patients with cervical cord compression at flexion and extension, significant differences [P<0.05] were found in patients with degenerative changes at 4 segments compared with patients with 1 segment involvement,” noted the authors.
In an enlarged subsequent analysis (n=81), many of these authors collaborated with esteemed radiologist Donald Resnick to develop a classification system correlating the kinematic (positional) MRI with the degree of CSM.5 In patients with progressive degenerative disease of the cervical spine (stage I-IV spinal stenosis varied between a score of 0 [normal diameter canal and anatomy] to 3 [anterior and posterior impingement they call a “pincer effect”]). In neutral, advancing degenerative stage (I-IV) correlated with a 3 score of spinal cord compression: stage I (46%), stage II (26%), stage III (54%), stage IV (87%). This pincer appearance was accentuated in extension and reduced in flexion. Further, they remarked: “On the assumption that a reliable diagnostic indicator of the genesis of CSM is the demonstration of cord compression at the site of the cord lesion, kinematic MRI showed functional cord impingement in 4 of 6 patients [with stage IV disease], whereas no cervical cord compression was seen at the neutral position.” The authors suggest that the usefulness of obtaining additional flexion and extension MRI sagittal views reveals greater functional cervical cord impingement in extension, as well as better recognition of unsuspected cord compromise, which becomes more likely with advancing degenerative spondylitic stage.
In 2003, Chen et al reported similar findings.6 For 62 patients with cervical degenerative disease, 31% demonstrated functional cord compression in extension compared to 3% in flexion. Degenerative stage (P<0.001) and a neutral canal diameter of ≤10 mm (P<0.037) were predictive of cord compression.
Upright MRI also demonstrated similar findings,7 with more readily apparent cord compression in extension, along with recognition of reduced angular mobility, particularly at C4/5 and C5/6, in a study of 459 patients with cervical degenerative disc disease.8
Analogously, Muhle and Resnick also reported in 1998 how position affected neuroforaminal patency in patients with cervical radiculopathy.9 Dynamic MRI demonstrated that ipsilateral rotation and lateral bending increased nerve root compression consistent with what clinicians might expect with a positive Spurling test.
Use in Rheumatoid Arthritis
Another application of dynamic MRI is in very advanced rheumatoid arthritis (RA), which can lead to a potentially fatal atlantoaxial dislocation or subluxation from progressive destruction/laxity of the transverse ligament. In flexion, the odontoid process may compress the upper cervical cord, which is clinically expressed as Lhermitte’s sign, and if sufficiently forceful, can cause death. Traditionally, radiographs (x-rays) have been employed to identify this condition, but flexion-extension MRI views also have been studied. Dedicated MRI positioning devices have been engineered for patients with RA,10 as they have for patients with degenerative diseases,11,12 and some authors suggest that this imaging technique should be the diagnostic “study of choice for the upper rheumatoid C-spine” for presurgical staging.13
Positional Spinal Cord Compression Linked With FM
In 2002, Heffez also advocated viewing the cervical cord more dynamically with MRI. His suggestion was presented in the context of sorting out another complex presentation, fibromyalgia (FM), at a small meeting of the National Fibromyalgia Research Association, in Portland, Oregon. His analysis found that symptoms of FM and cervical myelopathy overlapped for many patients with FM. Further, he noted that when he surgically decompressed severe, comorbid myelopathy, FM symptoms (allodynia, fatigue, central sensitization, pain, dyscognition, sleep disturbance) often abated.14 Similar observations were made for FM combined with Chiari malformation.15