| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Head and Neck Imaging |
1 From the Departments of Radiology (L.J.M.K., M.R., B.M.V., J.L.B., M.A.v.B.) and Rheumatology and Clinical Epidemiology (M.K.), Leiden University Medical Center, C2S, Albinusdreef 2, 2333 ZA Leiden, the Netherlands. Received December 9, 2002; revision requested February 6, 2003; final revision received July 22; accepted September 24. Address correspondence to L.J.M.K. (e-mail: l.j.m.kroft@lumc.nl).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Data from 33 consecutive patients with rheumatoid arthritis were evaluated; these patients had undergone 1.5-T magnetic resonance imaging following gadolinium chelate administration, in combination with a frequency selective fat-suppression technique. Stenosis and enhancement were scored for each of six cervical spinal levels and were compared with results in a control population consisting of 16 patients with degenerative disease. Enhancement was scored as superficial or deep on the anterior and posterior sides from the cervical spinal cord. Differences between patient groups were tested by using the 
2 test for trend and the Fisher exact test.
RESULTS: No significant difference was found in the frequency or severity of subaxial stenosis between rheumatoid arthritis and degenerative disease. Deep epidural enhancement was observed more often with rheumatoid arthritis than with degenerative disease both anterior (25 of 33 patients vs seven of 16 patients, respectively; P < .001) and posterior (24 of 33 patients vs two of 16 patients, respectively; P = .001) to the spinal cord. Enhancing stenosing tissue in rheumatoid arthritis frequently occurred anterior and posterior at the same time and at the same level, with segmental cufflike extension of enhancing tissue around the dural sac. Stenosing tissue enhanced more frequently with rheumatoid arthritis than with degenerative disease (22 of 33 vs four of 16 patients, respectively; P = .008).
CONCLUSION: In patients with rheumatoid arthritis, subaxial stenosis is frequently caused by enhancing epidural tissue. This enhancing tissue presumably represents pannus.
© RSNA, 2004
Index terms: Arthritis, rheumatoid, 34.71 • Spinal canal, stenosis, 34.71 • Spine, arthritis, 34.71 • Spine, MR, 34.121415, 34.12143
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
It is unknown whether subaxial spinal canal stenosis and subsequent neurologic dysfunction in rheumatoid arthritis are caused by active inflammatory tissue (pannus), degenerative changes, or both. If spinal canal stenosis is found to be caused by inflammatory tissue, there may be therapeutic consequences. Instead of surgery, which is currently the preferred treatment when neurologic symptoms and signs are present (4), focal conservative therapy (local application of steroids) might be considered to prevent disease progression.
In routine imaging of rheumatoid arthritis, we frequently encounter enhancement of soft tissue at subaxial stenosed levels on gadolinium-enhanced fat-suppressed magnetic resonance (MR) images. Thus, the purpose of our study was to assess the frequency and site of subaxial spinal canal stenosis due to enhancing soft tissue in patients with rheumatoid arthritis.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
There was no difference in the distribution by sex between patients with rheumatoid arthritis and those with degenerative disease. There were 22 women and 11 men with rheumatoid arthritis. The group with degenerative disease consisted of seven women and nine men. Among women, the mean age was 62 years (range, 40–79 years) for patients with rheumatoid arthritis and 54 years (range, 42–66 years) for patients with degenerative disease (P = .11). Among men, the mean age was 64 years (range, 41–81 years) for patients with rheumatoid arthritis and 53 years (range, 40–67 years) for patients with degenerative disease (P = .04).
MR Imaging Protocol
If more than one MR imaging examination was registered per patient, images from the first examination were used for analysis. MR imaging examinations were performed at 1.5 T (Gyroscan NT15; Philips Medical Systems, Best, the Netherlands) by following the standard protocol of our institution. Each series was obtained with a quadrature transmit/receive neck coil, with the patient supine and the neck in neutral position. The following sequences were used: (a) a sagittal T1-weighted fast spin-echo series with 500/7 (repetition time msec/echo time msec), echo train length of five, section thickness of 3.0 mm, field of view of 275 x 275 mm, rectangular field of view of 100% or 80%, three or four signals acquired per data line, and an acquisition matrix of 307 x 512; (b) a sagittal T2-weighted fast spin-echo series with 3,393–4,637/150, echo train length of 15 or 16, section thickness of 3.0 mm, field of view of 275 x 275 mm, rectangular field of view of 100% or 80%, four or six signals acquired, and an acquisition matrix of 255 or 251 x 512; and (c) a sagittal T1-weighted fast spin-echo spectral presaturation with inversion recovery (SPIR) sequence performed approximately 2–5 minutes after gadolinium chelate administration with 500/7 or 8, echo train length of five or four, section thickness of 3.0 mm, field of view of 275 x 275 mm, rectangular field of view of 100% or 80%, between two and four signals acquired, and an acquisition matrix of 307 x 512. The number of sections for the sagittal sequences was 11–12. The section gap was 0.3 mm.
In addition, transverse three-dimensional T2-weighted fast field-echo imaging was performed when indicated, as was the case in most patients, depending on findings of stenosis on the sagittal images. The imaging parameters of this sequence were as follows: 41/14, section thickness of 1.5 mm, 60 sections, field of view of 230 x 230 mm, rectangular field of view of 50%, two signals acquired, flip angle of 7.0, and an acquisition matrix of 212 x 256. We administered gadopentetate dimeglumine (Magnevist; Schering, Weesp, the Netherlands) intravenously by using manual injection, with a dose of 0.2 mL per kilogram of patient body weight.
Analysis of MR Images
The images were reviewed by two neuroradiologists (M.A.v.B. and B.M.V., 10 and 2 years experience in neuroradiology, respectively) who were unaware of clinical information or other patient data. In each case, consensus was obtained for each score.
All six subaxial cervical levels from C2–3 through C7-T1 were judged separately. Stenosis was defined as follows for severity: no stenosis, waisted (epidural tissue outlined beyond the normal margins of the vertebral bodies or dorsal elements), cord contact without cord deformation, or cord compression with cord deformation. Cord contact without cord deformation and cord compression with cord deformation were considered to be present only when the absence of cerebrospinal fluid on both sides of the spinal cord could be confirmed on the transverse MR images. Cord compression with cord deformation was defined as decreased cord diameter at the level of obliterated cerebrospinal fluid compared with the cord diameter superior or inferior to that stenosed level. The cause of stenosis was defined as discopathy (disk bulging or disruption and/or herniation), ligamentum flavum hypertrophy, spondylosis, or subluxation.
When reviewing the MR images, we observed two patterns of enhancement: superficial and deep enhancement. Enhancement was scored as superficial or deep on midsagittal and parasagittal MR images. Enhancement was seen on the anterior and posterior sides from the cervical spinal cord and was defined as a visually assessed increase of signal intensity on the postcontrast T1-weighted SPIR images, relative to the signal intensity on the precontrast T1-weighted images. No mass effect was observed with superficial enhancement. Superficial enhancement was defined as linear enhancement lining the cerebrospinal fluid, without involvement of deep structures other than the basivertebral vein. We assumed that this superficial enhancement would represent dura mater or epidural venous plexus. Deep enhancement was defined as enhancement that extended beyond the cerebrospinal lining. Enhancing deep structures that could be recognized on the anterior side were disk, bone, and soft tissue; on the posterior side, ligamentum flavum, bone, soft tissues of the interspinal ligaments, paraspinal soft tissue, ligamentum nuchae, and facet joints could be recognized. Furthermore, we assessed whether stenosing structures enhanced. Enhancement on stenosed levels was considered present only on levels with accompanying deep enhancement, as no mass effect was observed with superficial enhancement.
Diagnosis
Medical records from all included patients were retrieved by a rheumatologist (M.K.), and it was determined whether the patients were diagnosed as having rheumatoid arthritis as defined according to the 1987 criteria of the American Rheumatism Association (5). Rheumatoid arthritis was confirmed in all 33 patients according to these criteria. The duration of the disease was 1–47 years (median, 22 years). All patients with rheumatoid arthritis had erosive disease visible on hand-wrist and/or feet radiographs, and 30 of 33 patients were positive for rheumatoid factor.
Statistical Analysis
Differences in the frequency and cause of stenosis and differences in enhancement between rheumatoid arthritis and degenerative disease were tested by using the 2 test for trend to determine differences between patients. With this test, the six subaxial cervical spinal levels were clustered per patient by taking into account the frequencies of stenosis and enhancement in each patient. Calculations were then performed by using 2 x 7 and 2 x 4 tables for the distribution of enhancement of stenosing tissue. The Fisher exact test (2 x 2 tables) was used to determine enhancement of the ligamentum nuchae and facet joints. Analyses were performed by using a statistical software package (SPSS 10.07; SPSS, Chicago, Ill). For each test, the null hypothesis was rejected at 95% CIs. P < .05 was considered to indicate a significant difference.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
|
|
|
|
Enhancement on Stenosed Levels
Table 2 presents the frequency and sites of deep enhancement on stenosed levels. Enhancement on stenosed levels was much more frequent with rheumatoid arthritis than with degenerative disease. In particular, the occurrence of enhancement of anterior and posterior tissue at the same time on these stenosed levels was characteristic in patients with rheumatoid arthritis (14 of 33) but was not observed in any of the patients with degenerative disease (P = .006). In eight patients (11 stenosed levels) in the rheumatoid arthritis group, enhancement of the anterior and posterior structures was continuous with lateral extension of the enhancement in the paraspinal soft tissue and facet joints, which resulted in subaxial "cuff" enhancement (Fig 3). The severity of the stenosis at these levels was as follows: waisted (six occurrences), cord contact without cord deformation (two occurrences), and cord compression with cord deformation (three occurrences). Enhancement of at least one facet joint was observed in each of 15 patients with rheumatoid arthritis but in none of the patients with degenerative disease (15 of 33 vs none of 16, respectively; P = .001).
|
A marked observation was that stenoses were most often caused by discopathy in both groups: in 100 of 128 levels in the rheumatoid arthritis group and in 45 of 53 levels in the degenerative disease group. However, enhancement was seen in 49 of the 100 disks in the rheumatoid arthritis group, while enhancement was seen in only five of the 45 disks in the degenerative disease group.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
The frequency of superficial enhancement between stenosed and nonstenosed levels was comparable. We assume that superficial enhancement is a normal finding. Superficial enhancement could be explained by accumulation of contrast agent in the epidural venous plexus and/or by enhancing dura mater. The epidural space contains fat and many thin-walled veins (the internal vertebral venous plexus) (10). Below the foramen magnum, the dura mater is thick and vascular (11). With the currently available MR imaging techniques, it was not possible to distinguish epidural venous plexus and dura mater from each other. The finding that superficial enhancement can be observed more often on the anterior side than on the posterior side supports the hypothesis of epidural venous plexus enhancement, because the venous plexus is wider at the anterior side than at the posterior side (10).
Although the frequency of superficial enhancement was comparable, deep enhancement was observed more frequently on stenosed (diseased) levels than on nonstenosed levels. We therefore consider deep enhancement to be an indication of disease. Deep enhancement was observed much more frequently with rheumatoid arthritis than with degenerative disease. Stenoses were most often caused by discopathy in both groups, whereas enhancement was seen in 49 of 100 of these disks in the rheumatoid arthritis group but in only five of 45 of these disks in the degenerative disease group. This suggests a different cause of disk disease in rheumatoid arthritis than in degenerative disease. Even if degenerative disk disease is additionally present in rheumatoid arthritis, the high frequency of enhancement should be explained by other factors of rheumatoid arthritis; active inflammation seems a plausible cause.
The cause of disk destruction in rheumatoid arthritis has been well investigated. A rheumatoid cervical disk lesion is secondary to an erosive process arising in the neurocentral joints (also called the uncovertebral joints or Luschka joints). These joints are formed by a cleft in the lateral margin of the disk. The cleft is covered laterally by a fibrous membrane lined by synovium. Granulation tissue arises in these joints and spreads around the disk-bone border, gradually replacing the annulus (usually posteriorly at first) (2,6,12). Disk enhancement observed in our study was often present locally at the posterior disk site, and disks were also observed to enhance more extendedly or as a whole.
Disk enhancement in degenerative cervical spinal disease is attributed to granulation tissue within the annulus (annular tear) and to epidural fibrosis, which is associated with disk disruption and herniation (13,14). Disk disease may ultimately result in spinal canal stenosis in rheumatoid arthritis, as well as in degenerative disease.
If enhancement occurred on a stenosed level in a patient with rheumatoid arthritis, this enhancement was on both the anterior and posterior side of the spinal cord at the same time in almost half of the cases. This segmental enhancement was often observed to extend laterally, thereby connecting anterior and posterior enhancing areas with enhancement of paraspinal soft tissue (pannus) and enhancement of the facet joints. At histologic examination, rheumatoid arthritis lesions in facet joints closely resemble those in other synovial joints (2).
Although no histologic evaluation was performed in our patients, there is close similarity between findings in our study and findings described by investigators who examined cervical spines in rheumatoid arthritis histologically (15). The constricting epidural rings that led to cervical cord compression on subaxial levels, as observed in our patients with rheumatoid arthritis, were also found in a study by Kudo et al (15). At surgery, they observed a bandlike mass of soft tissue in the epidural space that formed a stenosing ring, which was present in several patients with rheumatoid arthritis. On a microscopic level, this ring consisted of degenerated ligamentum flavum and pannus. Furthermore, these stenosing rings coincided with erosive changes and marked proliferation of synovial tissue of the facet joints at the same levels in four of five patients. The description by Kudo et al at histologic evaluation matches precisely with our observations at radiologic evaluation. The cuff enhancement of stenosing tissue anterior and posterior from the spinal cord and the lateral extension into the facet joints on parasagittal MR images, therefore, suggest a focus of active inflammation.
Cuff enhancement and enhancement of the facet joint, paraspinal soft tissue, and ligamentum nuchae were observed in rheumatoid arthritis but not in degenerative disease. Although it was not the focus of our study, we mention the ligamentum nuchae enhancement in rheumatoid arthritis because, to our knowledge, this characteristic has not yet been reported in the literature. Given that enhancement in gadolinium-enhanced MR imaging has been shown to represent local disease activity in rheumatoid arthritis (16–18), it is very likely that deep tissue enhancement on subaxial levels in rheumatoid arthritis represents a focus of active inflammation (pannus and/or diseased synovium).
The frequency and location of enhancement in cervical spinal stenosis indicate that the cause of subaxial stenosis is different in rheumatoid arthritis than it is in degenerative disease. Our study results suggest that, in patients with rheumatoid arthritis, subaxial spinal canal stenosis with enhancement may represent a focus of active inflammation. The presence of active inflammation that gives rise to stenosis of the cervical canal may warrant different treatment than the currently used surgical approach in endstage stenosis; local medical treatment may be used to prevent disease progression at an earlier stage of disease. We have shown that images of these foci of active inflammation can be obtained by using a gadolinium-enhanced SPIR fat-suppression MR imaging technique. This technique can be used to evaluate disease activity in rheumatoid arthritis and may be helpful for planning and evaluating local conservative therapy.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Author contributions: Guarantor of integrity of entire study, L.J.M.K.; study concepts, M.A.v.B., J.L.B., M.R.; study design, M.A.v.B., M.R.; literature research, M.A.v.B., L.J.M.K.; clinical studies, M.A.v.B., M.R.; data acquisition, M.A.v.B., M.R.; data analysis/interpretation, M.A.v.B., B.M.V., L.J.M.K.; statistical analysis, L.J.M.K., M.K.; manuscript preparation, L.J.M.K.; manuscript definition of intellectual content and editing, L.J.M.K., M.A.v.B., J.L.B.; manuscript revision/review, M.R., M.K., B.M.V., J.L.B., M.A.v.B.; manuscript final version approval, all authors
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
