Are magnetic resonance flexion views useful in evaluating the cervical spine of patients with rheumatoid arthritis?

Objective: To determine whether MR imaging in flexion adds value relative to imaging in the neutral position with respect to displaying involvement of the subarachnoid space, brainstem and spinal cord.
Design and Patients: T1-weighted MR images of the cervical spine in 42 rheumatoid arthritis patients with cervical spine involvement were obtained and analyzed prospectively. We assessed changes between images obtained in the neutral position and following active flexion, especially horizontal atlantoaxial and subaxial motion, presence or absence of brainstem compression, subarachnoid space involvement at the atlantoaxial and subaxial level and the cervicomedullary angle. Vertical atlantoaxial subluxation and the amount of pannus were correlated with motion and change in subarachnoid space.
Results: The flexion images showed horizontal atlantoaxial motion in 21 patients and subaxial motion in one patient. The flexion view displayed brainstem compression in only one patient. Involvement of the subarachnoid space increased at the atlantoaxial level in eight (19%) patients (P=0.004) and at the level below C2 in five (12%) patients (P=0.03). There were no patients with a normal subarachnoid space in neutral position and compression in the flexed position. The cervicomedullary angle changed significantly with flexion. Vertical atlantoaxial subluxation and the amount of pannus did not show a significant correlation with motion or subarachnoid space involvement.
Conclusion: MR imaging in the flexed position shows a statistically significant narrowing of the subarachnoid space at the atlantoaxial level and below C2. Cord compression is only observed on flexion views if the subarachnoid space in neutral position is already decreased. MR imaging in the flexed position might be useful, since subarachnoid space involvement may be an indicator for the development of neurologic dysfunction.