Hidefumi Kawaida

Upper cervical involvement in rheumatoid arthritis.

Atlanto-axial dislocation and upward migration of the odontoid were studied in patients with rheumatoid arthritis treated conservatively and those treated by atlanto-axial posterior fusion. Upward migration of the odontoid was evaluated by Ranawats and Redlund-Johnells methods after determining the normal range in healthy Japanese adults. Of the 100 patients treated conservatively, atlantoaxial dislocation was found in 49% and upward migration of the odontoid in 26% by the former method, and in 8% by the latter method. The detection rate of these lesions increased with the increase in the duration of the disease and progression of rheumatoid lesions. Abnormal Ranawat values are associated with lesions in the C1–C2 segment and abnormal Redlund-Johnell values with 0–C2 lesions. The latter suggests severe conditions and seems to be a useful indicator for the diagnosis of upward migration of the odontoid in our study. Though rheumatoid lesions in the upper cervical spine mainly involved the C1–C2 level, marked deterioration of the atlanto-occipital joints was also occasionally observed. Therefore, in surgical treatment, attention should be paid also to this region. Patients with abnormal Redlund-Johnell values require spinal fixation including the occipital bone.

Magnetic resonance imaging of upper cervical disorders in rheumatoid arthritis

Upper cervical spine was examined with magnetic resonance imaging (MRI) and conventional roentgenograms in 55 patients with rheumatoid arthritis. The MRI findings were compared with various values determined in roentgenograms: the atlanto-dental interval (ADI), the space available for the spinal cord (SAC), and the Ranawat and Redlund-Johnell values. In patients with vertical settling (VS), MRI showed medullary compression in all those with abnormal Redlund-Johnell values and Ranawat values of 7 mm or less. In patients with anterior atlanto-axial subluxation, compression of the upper cervical cord was observed in all patients with SAC of 13 mm or less and many of those with ADI of 8 mm or greater. This study indicated that medullary compression can be estimated by these values determined in roentgenograms of the cervial spine.

Occipitoatlantoaxial fusion utilizing a rectangular rod

Atlantoaxial subluxation has been treated conventionally by Gallie posterior fusion. This technique, however, has disadvantages such as the frequent occurrence of pseudarthrosis, a high probability of relapse, and the necessity of long-standing strict external fixation until bone fusion. To overcome these problems, posterior occipitoatlantoaxial fusion was performed using a rectangular rod that assures strong internal fixation in 16 patients with atlantoaxial subluxation. The condition was complicated by superior migration of the dens in five patients. Clinical and roentgenographic examinations before and after the operation showed improvements in neurological symptoms and in pain in the neck and occipital region in all patients. Bone fusion was observed in all patients and reduction, performed to the extent possible during the operation, was retained adequately. The present method, which provides strong internal fixation, allows bone fusion and early initiation of rehabilitation with a simple external support of the neck. It also facilitates laminectomy of C1 in patients with associated myelopathy. This procedure, therefore, is particularly effective in patients with marked instability or with rheumatoid arthritis and makes postoperative application of a halo vest or skull traction unnecessary.

Vertical settling in rheumatoid arthritis. Diagnostic value of the Ranawat and Redlund-Johnell methods.

The usefulness of the recently developed Ranawat and Redlund-Johnell craniometric methods was compared with that of the conventional McGregor method for diagnosing vertical settling (VS) of the skull and the atlas on the axis in 209 patients with rheumatoid arthritis (RA). Statistical analysis of the values obtained from roentgenograms revealed close correlations among the three methods. The first two methods were superior to the McGregor method because the measuring points could be identified on plain roentgenograms of the cervical spine in virtually all cases. The McGregor value could not be determined in 38 (18%) patients. Since medullary compression could be detected by magnetic resonance imaging in all patients who showed abnormal Redlund-Johnell values, the Redlund-Johnell method may be useful for diagnosing advanced VS. It also detects not only atlantoaxial lesions but also atlantooccipital lesions. Thus, the Redlund-Johnell method appears to be the best method for diagnosing VS in RA patients.