| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Radiology, University of California, San Francisco, 505
Parnassus Ave., Box 0628, San Francisco, CA 94143-0628.
2 Department of Radiology, NYU Medical Center, 560 First Ave., TCH-HW202, New
York, NY 10016-6497.
3 Division of Rheumatology, University of California, San Francisco, 400
Parnassus Ave., ACC 587, Box 0326, San Francisco, CA 94143-0326.
Received June 18, 2003;
accepted after revision October 6, 2003.
Address correspondence to B. Taouli.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Eighteen patients with rheumatoid arthritis underwent conventional high-field-strength MRI, low-field-strength dedicated extremity MRI, and conventional radiography of both hands and wrists. Two independent reviewers searched for the presence and extent of bone erosions, joint-space narrowing, and synovitis. Bone erosions (E scores) and joint-space narrowing (J scores) were evaluated at 14 and 13 sites, respectively, on conventional high-field-strength MRI, low-field-strength dedicated extremity MRI, and radiography, using the Sharp-Genant scoring system. Synovitis (S scores) were evaluated at 13 sites on conventional high-field-strength MRI and low-field-strength dedicated extremity MRI.
RESULTS. For the detection of bone erosions, we found no significant difference (p = 0.71) between conventional high-field-strength MRI (mean ± SD E score, 27.5 ± 9.8) and low-field-strength dedicated extremity MRI (28.8 ± 10.0), but a significant difference (p < 0.001) appeared between MRI and radiography (13.1 ± 8.3). J scores derived from MRI (conventional high-field-strength MRI, 15.2 ± 8.3; low-field-strength dedicated extremity MRI, 14.5 ± 10.4) were higher than those derived from radiography (12.7 ± 9.6), although the difference was not significant (p = 0.70). Conventional high-field-strength MRI (S score, 35.1 ± 8.6) and low-field-strength dedicated extremity MRI (30.8 ± 10.2) were equivalent (p = 0.14) for the evaluation of synovitis. The interobserver agreement for MRI scores was good to excellent (intraclass correlation coefficients, 0.83–0.94).
CONCLUSION. Conventional high-field-strength MRI and low-field-strength dedicated extremity MRI showed similar results in terms of cross-sectional grading of bone erosions, joint-space narrowing, and synovitis in the hands and wrists of patients with rheumatoid arthritis.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The purpose of our study was to compare results from conventional high-field-strength 1.5-T MRI, low-field-strength dedicated extremity 0.2-T MRI, and radiography to detect and grade bone erosions, joint-space narrowing, and synovitis in the hands and wrists of patients with rheumatoid arthritis.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Imaging Procedures
All patients underwent high-field-strength 1.5-T conventional MRI (Signa,
General Electric Medical Systems), low-field-strength 0.2-T dedicated
extremity MRI (Artoscan, Esaote Biomedica), and radiography of both hands and
wrists on the same day.
Radiography.—Posteroanterior images of both hands were obtained using a conventional technique (single-emulsion, single-screen, high-resolution system). The exposure aimed to achieve optimal visualization of the trabeculae and joints: small focal spot, 40 inches (102 cm); focus–field distance, 50–55 kV; 100 mA at 300 msec.
Conventional high-field-strength MRI and low-field-strength dedicated extremity MRI.—Conventional high-field-strength MRI and low-field-strength dedicated extremity MRI were performed on four areas: right and left wrists and right and left fingers. No IV contrast material was used.
On conventional high-field-strength MRI, the following sequences were obtained using a commercial circumferential wrist coil with a 3-inch internal diameter: coronal T1-weighted spin-echo (TR/TE, 600/9; matrix, 512 x 192; field of view, 12 x 12 cm; number of acquisitions, 2; slice thickness, 3 mm; acquisition time, 3 min 30 sec) on both sides, and coronal 3D T2* fast gradient-echo sequences in steady state with frequency-selective fat saturation (29.4/6.3; matrix, 512 x 192; flip angle, 20°; field of view, 12 x 12 cm; number of acquisitions, 2; slice thickness, 1.5 mm; acquisition time, 6 min) on the dominant hand. The choice of the MRI sequences and parameters was aimed at achieving a balance among anatomic coverage, contrast-to-noise ratio, spatial resolution (using the fast gradient-echo sequence), patient tolerance, and technical simplicity.
On low-field-strength dedicated extremity MRI, the following sequences were obtained: coronal T1 isotropic 3D gradient-recalled echo sequences (30/12; matrix, 192 x 160; flip angle, 60°; field of view, 15 x 12 cm; number of acquisitions, 2; slice thickness, 3 mm; acquisition time, 5 min 30 sec) on both sides, and coronal short time inversion recovery sequences (1,000/16; inversion time, 80 msec; matrix, 192 x 136; field of view, 18 x 12 cm; number of acquisitions, 3; slice thickness, 3.5 mm; acquisition time, 6 min 6 sec) on the dominant hand.
The total acquisition time (not including the set-up time) was 26 min for conventional high-field-strength MRI and 34 min for low-field-strength dedicated extremity MRI.
Image Evaluation
Seventy-two MRI examinations and 36 radiographs were evaluated in consensus
by two trained observers on three separate occasions. Results from
conventional high-field-strength MRI, low-field-strength dedicated extremity
MRI, and radiography were evaluated separately—first independently and
then in consensus.
The following parameters were assessed: bone erosions and joint-space narrowing on MRI and radiography; synovial hypertrophy and joint effusion together on MRI.
The Genant-modified Sharp radiographic scoring method [12, 13] was used for radiography and adapted for MRI (Table 1). This scoring method was initially developed for radiography and has been successfully used in clinical trials.
|
Radiographic evaluation.—Bone erosions and joint-space narrowing were scored on each radiograph on a paper form using the previously validated method [13] according to the grading system described in the next section. Independent scores for erosions and joint-space narrowing were obtained for each hand and wrist. The intraobserver and interobserver reproducibility of this method was reported previously [13].
Erosion score (E score).—Bone erosions were evaluated at 14 sites in each hand and wrist (including the proximal interphalangeal joints and the metacarpophalangeal joints, the first carpometacarpal joint, the scaphoid bone, and the distal parts of the ulna and the radius), using an 8-point scale from 0 to 3.5 based on the number and size of erosions and the area of bone involved: 0 (normal, no erosions), 0.5 (subtle change, subtle loss of cortical continuity or equivocal findings of bone erosion), 1.0 (mild, definite but small erosions of one or both articular bones, usually at the bare areas, typically involving < 25% of the combined articular surface area), 1.5 (mild to moderate, small to medium erosions usually involving < 25% of the articular surface of one or both articular bones), 2.0 (moderate, medium to large erosions involving as much as 50% of the articular surface of one or both articular bones), 2.5 (moderate to severe, erosion of 50–75% of the articular surfaces), 3.0 (severe, erosion of > 75% but not all articular surfaces), and 3.5 (severe worse, erosion of the entire articular surface).
Joint-space narrowing score (J score).—Joint-space narrowing was evaluated at 13 sites in each hand and wrist. Sites included the proximal interphalangeal joints and the metacarpophalangeal joints, the carpometacarpal joints three to five as a single unit, the pericapitate space (scaphoid–capitate and lunate–capitate combined), and the radiocarpal joint. We used a 5-point scale from 0 to 4: 0 (normal joint space), 1 (mild, narrowing of the joint space width), 2 (moderate, narrowing of the joint space width), 3 (severe, nearly complete loss of the joint space), and 4 (osseous ankylosis or dislocation).
MR image evaluation.—The MR images were evaluated on a workstation using modified version 1.6 software (MRVision). Erosions were defined as focal, sharply marginated defects that replaced normal fatty marrow and cortical and trabecular bone and were continuous with the bone surface. Bone erosions were distinguished from marrow edema principally on the basis of margin sharpness. Erosions were considered to have sharp, well-defined margins, and edema had ill-defined margins. Erosions and joint-space narrowing were scored independently according to the same grading scheme used for the radiographs. Independent scores for erosions and joint-space narrowing were obtained for each hand and wrist.
Synovial hypertrophy score (S score).—Synovial hypertrophy
and joint effusion were defined as the presence of interarticular thickened
synovium with or without fluid that appeared hypointense on T1- or
hyperintense on T2-weighted images (dominant hand) and as hypointense on
T1-weighted images (nondominant hand) and were evaluated together as a single
measure of synovitis on a 4-point scale from 0 to 3 at 13 sites (the proximal
interphalangeal joints and the metacarpophalangeal joints, the first
carpometacarpal joint, all wrist joints as a single unit, and the distal
radioulnar joint) according to the following scoring system: 0 (none), 1
(mild, > one third maximal potential joint distention), 2 (moderate, 
two thirds maximal potential joint distention), 3 (severe, > two thirds
maximal potential joint distention).
Statistical Evaluation
The total E scores and J scores obtained from conventional
high-field-strength MRI, low-field-strength dedicated extremity MRI, and
radiography; and S scores from conventional high-field-strength MRI and
low-field-strength dedicated extremity MRI were obtained for each patient by
consensus evaluation and compared using the Wilcoxon's signed rank test and
the analysis of variance test (two-sample rank sum test) where appropriate. A
p value less than 0.05 was considered significant. The interobserver
agreement on MRI was evaluated using the intraclass correlation coefficients
computed from the independent evaluation results.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Bone Erosions
Bone erosions were generally more conspicuous and well delineated on MR
images than on radiographs. The E scores derived from low-field-strength
dedicated extremity MRI and high-field-strength MRI were not statistically
different (Table 2 and Figs.
1A,
1B,
1C,
1D and
2A,
2B,
2C,
2D). MRI showed more erosions
than radiography, with mean E scores more than twice as high on MRI as on
radiography.
|
|
|
|
|
|
|
|
|
Joint-Space Narrowing
The J scores derived from conventional high-field-strength MRI and
low-field-strength dedicated extremity MRI were not statistically different
(Table 2 and Fig.
3A,
3B,
3C,
3D). Joint-space narrowing was
rated as more moderate on radiography than it was on MRI, but the difference
was not statistically significant.
|
|
|
|
Synovitis
The S scores derived from conventional high-field-strength MRI and
low-field-strength dedicated extremity MRI were equivalent
(Table 2 and Figs.
2A,
2B,
2C,
2D and
4A,
4B,
4C,
4D).
|
|
|
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The major drawbacks of MRI are its high cost and limited availability. Dedicated low-field-strength MRI devices using less than 1 T offer advantages over high-field-strength systems: lower cost, easier siting, better patient comfort, and reduced patient risk [24]. A few previous studies have compared the usefulness of low-field-strength dedicated extremity MRI versus conventional high-field-strength MRI and have shown equivalent performances for the evaluation of shoulder lesions [25] or lower performances for the ankle and foot [26]. However, low-field-strength dedicated extremity MRI has the drawbacks of lower signal-to-noise ratio [27–29], limited anatomic coverage, and longer acquisition time compared with conventional high-field-strength MRI.
In rheumatoid arthritis, most of the previous MRI studies were carried out using high-field-strength systems [4, 5, 7, 8, 30–32]. A few studies have used low-field-strength systems [9–11, 33] and have reported better results than those from radiography. However, only one study has directly compared conventional high-field-strength MRI with low-field-strength dedicated extremity MRI in patients with various forms of arthritis, including rheumatoid arthritis, and it showed no difference in the measurement of synovial membrane volume after gadolinium injection or in the evaluation of bone erosions and bone edema in hand arthritis [9]. That study also found that 64% of patients said that low-field-strength dedicated extremity MRI was more comfortable than conventional high-field-strength MRI when they responded to a questionnaire with a 4-point scale evaluating the discomfort from each MRI technique. Our study showed that low-field-strength dedicated extremity MRI and conventional high-field-strength MRI performed equally well for cross-sectional grading of bone erosions, joint-space narrowing, and synovitis in rheumatoid arthritis. Both MRI techniques detected approximately twice as many erosions as radiography did. They were more sensitive than radiography for the evaluation of joint-space narrowing, although no significant difference appeared, possibly as a result of the small sample size.
Evaluation of Bone Erosions in Rheumatoid Arthritis
A direct link between synovitis and bone damage is still controversial, but
strong evidence exists that early bone changes, such as bone edema, rarely
occur in the absence of synovitis. Several authors have suggested a
MRI-tracked three-step sequence of changes from synovitis to bone edema and
then to erosions [7,
34]. Patients with a
pronounced carpal synovitis at baseline were more at risk than others to
develop erosions at 1 year
[23,
35] or 2 years
[36]. Direct pathologic
confirmation that the erosions seen on MRI represent the same structural
lesions that are depicted on radiographs has yet to be reported, but
considerable indirect evidence supports this contention. Several studies
[5,
7,
37] have reported a higher
prevalence of marrow edema in rheumatoid arthritis and have presented evidence
to suggest that edema and inflammation in the bone marrow may precede the
development of actual bone erosions. However, the patients in those studies
had early active disease (< 6 months) and showed substantial progression of
erosive damage, and our study examined patients with chronic rheumatoid
arthritis on stable therapy.
Further evidence that the bone erosions depicted on MR images are the same lesions that are seen on radiography comes from the study of Foley-Nolan et al. [21], which showed that all erosions detected on radiography were also visible on MRI and that the distribution of the lesions depicted by the two techniques correlated anatomically. In a longitudinal study of patients undergoing active therapy, Ostergaard et al. [6] reported that erosions visible on radiography after 1 year could be seen on MRI at least 6 months earlier and that the lag time between the appearance of erosions on MRI and their emergence on radiography was typically more than 1 year. McQueen et al. [5] also reported that erosions were visible on MRI 6–12 months before they appeared on radiography.
Evaluation of Joint-Space Narrowing in Rheumatoid Arthritis
Our study is the first, to our knowledge, to compare the evaluation of
cartilage damage and subsequent joint-space narrowing between MRI and
radiography. We found that MRI was more sensitive than radiography, although
not significantly so, possibly because of the sample size. We found good
interobserver agreement for the evaluation of joint-space narrowing. However,
a low to moderate interobserver agreement for the evaluation has been reported
recently [38], and the
difficulty of differentiating cartilage from synovitis and joint fluid on MRI
in small joints must be emphasized.
Evaluation of Synovitis in Rheumatoid Arthritis
In our study, synovitis was visible on both T1-weighted spin-echo and T2*
fast gradient-echo sequences (conventional high-field-strength MRI) and on
coronal T1-weighted 3D gradient-recalled echo and short time inversion
recovery sequences (low-field-strength dedicated extremity MRI) as fluid
signal distending the joint cavity. Synovial tissue could not be reliably
differentiated from synovial fluid with the sequences used in this study.
Discriminating these two components of the synovial cavity typically requires
the use of IV contrast material or special sequences such as magnetization
transfer [39]. The injection
of gadolinium would have increased the complexity, cost, and duration of the
examination. Moreover, diffusion of the contrast agent from the synovium into
the adjacent joint fluid is known to occur rapidly in small joints and would
obscure the synovium. Accordingly, it was decided not to attempt to
discriminate between the synovium and synovial fluid, particularly because
both were essentially features of the same process of synovitis.
Limitations
Several limitations of our study should be mentioned. First, only a small
number of patients were evaluated, which limits the power of the statistical
results. However, a fair number of MRI examinations (n = 72) were
evaluated. Second, we did not use contrast injection for the evaluation of
synovitis, to limit the cost and duration of the examination. Third, we used
different T2-weighted sequences for conventional high-field-strength MRI and
low-field-strength dedicated extremity MRI. We used a T2* fast gradient-echo
in steady-state sequence for conventional high-field-strength MRI to obtain
high-resolution images, and a STIR sequence for low-field-strength dedicated
extremity MRI, because of the field strength limitation. However, the
comparison was based on an overall evaluation including both T1- and
T2-weighted images. Fourth, we did not evaluate bone marrow edema because most
of our patients had chronic rheumatoid arthritis and were undergoing stable
therapy.
Summary
Based on our results, despite its lower image quality and longer
acquisition time, low-field-strength MRI compares favorably with
high-field-strength MRI in the detection and grading of bone erosions,
joint-space narrowing, and synovitis in the hands and wrists of patients with
rheumatoid arthritis.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  J E Freeston, A K Brown, E M A Hensor, P Emery, and P G Conaghan Extremity magnetic resonance imaging assessment of synovitis (without contrast) in rheumatoid arthritis may be less accurate than power Doppler ultrasound Ann Rheum Dis, September 1, 2008; 67(9): 1351 - 1351. [Full Text] [PDF]
|
|
|
|
|
|
C. Schirmer, A. K Scheel, C. E Althoff, T. Schink, I. Eshed, A. Lembcke, G.-R. Burmester, M. Backhaus, B. Hamm, and K.-G. A Hermann Diagnostic quality and scoring of synovitis, tenosynovitis and erosions in low-field MRI of patients with rheumatoid arthritis: a comparison with conventional MRI Ann Rheum Dis, April 1, 2007; 66(4): 522 - 529. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H M Lindegaard, J Vallo, K Horslev-Petersen, P Junker, and M Ostergaard Low-cost, low-field dedicated extremity magnetic resonance imaging in early rheumatoid arthritis: a 1-year follow-up study Ann Rheum Dis, September 1, 2006; 65(9): 1208 - 1212. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Schoellnast, H. A. Deutschmann, J. Hermann, G. J. Schaffler, P. Reittner, F. Kammerhuber, D. H. Szolar, and K. W. Preidler Psoriatic arthritis and rheumatoid arthritis: findings in contrast-enhanced MRI. Am. J. Roentgenol., August 1, 2006; 187(2): 351 - 357. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A K Scheel, K-G A Hermann, S Ohrndorf, C Werner, C Schirmer, J Detert, M Bollow, B Hamm, G A Muller, G R Burmester, et al. Prospective 7 year follow up imaging study comparing radiography, ultrasonography, and magnetic resonance imaging in rheumatoid arthritis finger joints Ann Rheum Dis, May 1, 2006; 65(5): 595 - 600. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B J Ejbjerg, E Narvestad, S Jacobsen, H S Thomsen, and M Ostergaard Optimised, low cost, low field dedicated extremity MRI is highly specific and sensitive for synovitis and bone erosions in rheumatoid arthritis wrist and finger joints: comparison with conventional high field MRI and radiography Ann Rheum Dis, September 1, 2005; 64(9): 1280 - 1287. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
