| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Musculoskeletal Imaging |
1 From the Departments of Musculoskeletal Radiology (N.B., A.C.), Internal Medicine (E.H.), and Rheumatology (R.M.F., B.C.) and the Unité de Recherche de l'Appareil Locomoteur (N.B., R.M.F., B.C., A.C.), Hôpital Roger Salengro, Centre Hospitalier Régional Universitaire de Lille, Blvd du Professeur Leclercq, 59037 Lille CEDEX, France. Received May 10, 2004; revision requested July 27; revision received August 13; accepted September 11. Address correspondence to N.B. (e-mail: nboutry{at}chru-lille.fr).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MATERIALS AND METHODS: This study had institutional review board approval, and patient informed consent was obtained. Twenty-eight patients (16 female and 12 male patients; mean age, 42 years) with early RA and 19 patients (18 female and one male patient; mean age, 46 years) with SLE (n = 14) or primary Sjögren syndrome (n = 5) underwent MR imaging of both hands. All patients had inflammatory polyarthralgia of the hands and no evidence of erosive changes on radiographs. Coronal T2-weighted short inversion time inversion-recovery, transverse T1-weighted spin-echo, transverse fat-suppressed gadolinium-enhanced T1-weighted spin-echo, and transverse gadolinium-enhanced three-dimensional gradient-echo MR images were obtained. The following MR imaging variables were assessed in the wrist and nonthumb metacarpophalangeal joints: synovitis, bone lesions (erosion, defect, and edema), and tenosynovitis. Synovitis and bone lesions were scored with the OMERACT RA-MRI scoring system. Findings in patients with RA and those without RA were compared by means of Mann-Whitney, 
2, and Fisher exact tests.
RESULTS: The only significant difference between the two groups in terms of individual scores for synovitis, bone lesions, and tenosynovitis was the more frequent presence of tenosynovitis of the right fourth extensor tendon in patients without RA (P = .04). There were no significant differences between patients with RA and those without RA in terms of global scores for synovitis, bone lesions, and tenosynovitis. However, bone marrow edema in the metacarpophalangeal joints was seen more frequently in patients with RA (P < .001).
CONCLUSION: It may be impossible to distinguish between patients with early RA and those without RA (ie, those with SLE or primary Sjögren syndrome) by means of MR imaging.
© RSNA, 2005
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
The purpose of this study, therefore, was to evaluate prospectively the use of MR imaging for differentiating true RA from SLE or primary Sjögren syndrome in patients who have inflammatory polyarthralgia of the hands but no radiographic evidence of RA.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MR Imaging Protocol
MR imaging was performed with a 1.5-T superconducting magnet (Vision; Siemens; Erlangen, Germany) equipped with a transmit-receive, 20-cm-diameter circular surface coil. Both hands were imaged. The patients underwent imaging in a prone or supine position, with arms semiflexed above the head and hands positioned in the center of the coil. Straps kept the palms facing each other in "prayer position" and the fingers extended, aligning the carpus with the metacarpals.
In all patients, the imaging protocol began with coronal T2-weighted short inversion time inversion-recovery (STIR) and transverse T1-weighted spin-echo (SE) imaging. Coronal images were obtained parallel to the great axis of the carpus and the metacarpals. Transverse images were obtained perpendicular to the great axis of the carpus and the metacarpals. After intravenous bolus injection of 0.1 mmol per kilogram of body weight of gadopentetate dimeglumine (Dotarem; Guerbet, Roissy, France), fat-suppressed gadolinium-enhanced transverse T1-weighted SE and gadolinium-enhanced three-dimensional transverse fast low-angle shot (FLASH) images were obtained.
For T2-weighted STIR imaging, the parameters were as follows: 5216/60/150 (repetition time msec/echo time msec/inversion time msec); flip angle, 180°; field of view, 23 x 23 cm; section thickness, 4 mm; two signals acquired; matrix, 286 x 512; and acquisition time, 4 minutes 36 seconds. Sixteen sections were obtained with this sequence. For T1-weighted SE imaging, the parameters were as follows: 570/20 (repetition time msec/echo time msec); field of view, 15 x 20 cm; section thickness, 4 mm; one signal acquired; matrix, 379 x 512; and acquisition time, 3 minutes 39 seconds. The in-plane resolution was 0.4 x 0.6 mm. Sixteen sections were obtained with this sequence both before and after contrast agent administration. The imaging parameters used after intravenous bolus injection of gadopentetate dimeglumine—except repetition time (855 msec) and acquisition time (5 minutes 27 seconds)—were similar to those used before injection.
The FLASH sequence employed frequency-selective water excitation. A 45-mm slab was partitioned into 60 transverse sections, resulting in a section thickness of 1.5 mm. Other imaging parameters were as follows: 36/9; flip angle, 50°; field of view, 20 x 20 cm; one signal acquired; matrix, 300 x 512; and acquisition time, 5 minutes 25 seconds. The in-plane resolution was 0.4 x 0.7 mm.
With each transverse series, two stacks of sections (eight sections for the SE and 30 sections for the FLASH sequences) were obtained through the wrist and metacarpophalangeal (MCP) joints. The first series covered the wrists from the distal radioulnar joints to the metacarpal bases, and the second series covered the MCP joints. The overall examination time was about 20 minutes.
Follow-up Procedures and Final Study Population
Clinical follow-up by two of the authors (E.H., R.M.F.) was continued until a definitive diagnosis was made. The average length of follow-up, from the first to the final clinical visit, was 882 days (range, 124–2196 days). Fifty patients with a confirmed diagnosis of RA (n = 29), SLE (n = 16), or primary Sjögren syndrome (n = 5) were enrolled in the present study. Five of the 56 patients clinically suspected of having early RA had a confirmed diagnosis of reactive arthritis (n = 2) or unclassified self-limited arthritis (n = 3). One patient was lost to follow-up.
MR examination results in three patients could not be evaluated due to movement artifacts (n = 2) or an incomplete data set (n = 1). These patients were excluded from the cohort, leaving 47 patients with inflammatory polyarthralgia of the hands and a confirmed diagnosis of RA (n = 28), SLE (n = 14), or primary Sjögren syndrome (n = 5). With respect to the diagnosis of RA, patients were divided into two groups.
The 28 patients who fulfilled the 1987 American College of Rheumatology criteria for the diagnosis of RA (12) were included in the first group (12 male and 16 female patients; mean age, 42 years; age range, 17–69 years). The MR imaging examination was performed an average of 5 months (range, 2–7 months) after the onset of the first articular symptoms. Medication at entry included nonsteroidal antiinflammatory drugs in 16 patients and corticosteroids in five; none of these patients had previously received disease-modifying antirheumatic drugs. Seven patients were completely untreated.
The 14 patients who fulfilled the 1982 American College of Rheumatology criteria for the classification of SLE (13) (updated in 1997 [14]) and the five who fulfilled the criteria for the diagnosis of primary Sjögren syndrome (15) were included in the second group (one male and 18 female patients; mean age, 46 years; age range, 28–67 years). MR imaging in this group was performed an average of 9 months (range, 4–15 months) after the onset of the first articular symptoms. Medication at entry included nonsteroidal antiinflammatory drugs in eight patients and corticosteroids in two; none of these patients had previously received disease-modifying antirheumatic drugs. Nine patients were completely untreated.
Analysis of MR Images
Two radiologists (N.B., A.C.) who were blinded to the final diagnosis (ie, RA or other diagnosis) reviewed MR images in consensus for the presence of synovitis, tenosynovitis, and bone lesions. The observers were musculoskeletal radiologists with 10 and 15 years of experience with musculoskeletal MR imaging, respectively. Synovitis and bone lesions were evaluated according to the OMERACT RA-MRI scoring system (16).
Scoring of synovitis.—Synovitis was demonstrated by gadolinium enhancement of the synovial compartment on fat-suppressed gadolinium-enhanced T1-weighted SE MR images. A global score scale of 0–3 was used. A score of 0 indicated a normal synovial compartment with no synovial enhancement or an area of enhancement that was no thicker than the joint capsule. Scores of 1, 2, and 3, respectively, indicated areas of mild, moderate, or severe enhancement that were thicker than the joint capsule. These global scores were assigned to each of four MCP joint sites (the first MCP joint was not taken into account), yielding an aggregated score of 0–12 for each side (ie, right and left). In the wrist, the global score was assigned at three sites: the radioulnar, radiocarpal, and intercarpal-carpometacarpal joints, yielding an aggregated score of 0–9 for each side.
Scoring of bone lesions.—Bone lesions were scored from 0 to 10 in 10% increments according to the volume of the erosion as a proportion of the assessed bone volume and judged on all available images. For the carpal bones, the assessed bone volume was that of the whole bone. For the MCP joints (second to fifth), the assessed bone volume included a bone portion 1 cm deep from the cortex of the articular surface or the best estimated position of the cortex of the arterial surface if this cortex was absent (because of an erosion). Erosions were defined as sharply marginated areas of trabecular bone loss with a cortical break. Bone defects were defined as sharply marginated areas of trabecular bone loss without a visible cortical break. Bone edema could occur alone or surround a defect or an erosion and was defined as a lesion with ill-defined margins that was neither an erosion nor a defect and had high signal intensity on T2-weighted STIR MR images. For the MCP joints, the locations of bone abnormalities (ie, involvement of radial, ulnar, palmar, and dorsal quadrants) were also specified.
Bone abnormalities (erosions, defects, and edema) were assigned to 14 sites (distal portion of radius; distal portion of ulna; scaphoid, lunate, triquetrum, pisiform, trapezium, trapezoid, capitate, and hamate bones; and the bases of the second through fifth metacarpal bones) for each side on the wrist joint images and to eight sites (metacarpal head and phalangeal base of each second through fifth MCP joint) for each side on the MCP joint images. Therefore, the aggregated scores for wrist and MCP joints ranged from 0 to 140 and from 0 to 80, respectively, for each side with respect to bone erosions, bone defects, and bone edema.
Scoring of tenosynovitis.—Fat-suppressed gadolinium-enhanced T1-weighted SE MR images were used to assess tenosynovitis, which was considered to be present when enhancement was seen in the tendon sheath. A global score of 0–3 was assigned to tendon sheath enhancement. Grading was as follows: grade 0 indicated no tendon sheath enhancement; grade 1, tendon sheath enhancement without tendon sheath thickening; grade 2, tendon sheath enhancement with moderate tendon sheath thickening; and grade 3, tendon sheath enhancement with marked tendon sheath thickening. In the wrist, six tendon groups (the extensor pollicis longus and brevis, extensor carpi radialis longus and brevis, extensor pollicis longus, extensor digitorum, extensor digiti minimi, and extensor carpi ulnaris groups) were scored at the dorsal side of the joint, and two tendon groups (the flexor carpi radialis and flexor digitorum profundus and superficialis groups) were scored at the palmar side, yielding an aggregated score of 0–24 for each side. In the MCP, eight tendon groups (the second through fifth extensor and flexor tendons) were scored at both the dorsal and palmar sides of the joints, yielding an aggregated score of 0–24 for each side.
Other MR abnormalities.—Extracapsular changes (ie, regions of increased signal intensity on coronal T2-weighted STIR images or transverse fat-suppressed gadolinium-enhanced T1-weighted images in adjacent soft tissues) were also assessed.
Statistical Analysis
Statistical computations were performed with Statview 5 software (SAS Institute, Cary, NC). The data were first analyzed individually according to side, joint, and lesion (eg, synovitis of the right intercarpal-carpometacarpal joints, erosion of the left second MCP joint) and then analyzed as aggregated scores (eg, global wrist joint synovitis score, global MCP joint erosion score). Aggregated scores are presented as means, with ranges in parentheses. A Mann-Whitney U test for unpaired, nonparametric data was used to compare MR imaging findings of the hands in patients with RA with those in patients without RA. P values less than .05 were considered to indicate significant differences. For each type of MCP joint lesion (ie, erosion, defect, or edema), a 2 test was used to compare patients with and patients without RA in terms of the quadrant involved (ie, radial, ulnar, palmar, or dorsal). Finally, a Fisher exact probability test was used to compare the two groups in terms of the percentages of patients who had synovitis, bone lesions, or tenosynovitis.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
|
|
|
|
|
Bone Lesions
Bone lesions were found in patients with and in those without RA (Figs 1, 2). No significant difference was found between patients with and those without RA in terms of bone lesion scoring (ie, erosion, defect, and edema), although bone erosions and bone edema were seen more frequently in patients with than in those without RA (Table 1). In the wrist joints, bone erosions were seen in all 28 patients (100%) with RA and in 16 (84%) of 19 patients without RA, and bone edema was seen in 11 (39%) of 28 patients with RA and in three (16%) of 19 patients without RA. In the MCP joints, bone erosions were seen in 17 (61%) of 28 patients with RA and in nine (47%) of 19 patients without RA, and bone edema was seen in 20 (71%) of 28 patients with RA and in one (5%) of 19 patients without RA. However, only bone marrow edema within the MCP joints was significantly (P < .001) more frequent in patients with RA than in those without RA. Patients with primary Sjögren syndrome exhibited fewer bone lesions in the wrist joints than patients with early RA or SLE and exhibited no bone lesions in the MCP joints (Fig 3; Tables 1, 2). Except for bone edema in the wrist joints, the capitate bone and the third metacarpal head had highest global scores in both patients with and patients without RA (Table 3).
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Recently, an MR imaging system for scoring inflammatory and destructive changes in the wrist and MCP joints of patients with early RA was developed by the OMERACT MRI study group and subsequently tested for interreader agreement among several international centers (17,18). These preliminary results suggested that this MR imaging scoring system, also known as the RAMRIS system, had acceptable interreader reliability for measurements of disease activity and damage (global synovitis, bone edema, and bone erosion scores) (18). In the present study, we used the RAMRIS system as described elsewhere (19), but we also scored the MCP joint quadrants (ie, the radial, ulnar, palmar, and dorsal quadrants) and tenosynovitis to identify any potential differences between patients with and those without RA.
Active synovitis was seen on MR images in patients with and those without RA in all cases. Tenosynovitis was more frequently seen in patients with RA (82%) than in patients without RA (63%), although no significant difference was noted between the two groups. Such results are not surprising, since contrast enhancement within the synovial compartment or the tendon sheath is nonspecific for RA and has been described in various rheumatic diseases (2,20–25). Sugimoto et al (1) defined active synovitis as bilateral periarticular enhancement of the wrist or MCP and/or proximal interphalangeal joints. They showed that both sensitivity and specificity were improved when this MR imaging criterion was incorporated into the classification tree format of the 1987 American Rheumatism Association revised criteria for the classification of RA (sensitivity improved from 69% to 100%, and specificity improved from 64% to 73%). However, three false-positive diagnoses were encountered in patients who had SLE with cryoglobulinemia, arthritis related to viral infection, or Behçet syndrome. Active inflammation in the joints and subsequent increased vascular permeability from any cause may be responsible for gadolinium diffusion and periarticular enhancement. In the present study, we found tenosynovitis of the right fourth extensor tendon to be the only MR variable that differed significantly between patients with RA and those without RA. However, owing to the large number of statistical tests performed when MR variables were considered individually, this result has to be interpreted carefully.
On MR images, bone erosions and bone edema were identified more frequently in patients with early RA than in those without RA, although no significant differences were found in global or individual scores. Not finding any significant differences in terms of bone lesions between patients with and those without RA was more surprising than not finding any significant differences in terms of synovitis and tenosynovitis between patients with and those without RA (because erosive bone changes are usually considered to be specific for RA). To date, little information is available on MR imaging features of the hands in patients without RA (20–22,24); to our knowledge, only one report has described the MR findings in SLE (11) and only one article has cited primary Sjögren syndrome among a variety of rheumatic conditions (21).
Ostendorf et al (11) used MR imaging to evaluate 14 patients with both SLE of varying severity and duration and arthritis of the hand and finger joints. A persistent hand deformity known as Jaccoud arthropathy was also present in eight (57%) of the 14 patients. MR imaging abnormalities such as capsular swelling (all patients), active synovitis (nine patients [64%]), tenosynovitis (10 patients [71%]), and bone erosions (eight patients [57%]) were identified. Bone erosions were found predominantly at the MCP joints (eight patients [57%]) but also at the wrist joints (seven patients [50%]). Patients with a deforming arthropathy showed a tendency to have edematous capsular swelling and tenosynovitis without marked synovial membrane contrast enhancement and alterations in the bone, while patients without a deforming arthropathy showed a tendency to have synovial membrane contrast enhancement (ie, active synovitis) and bone alterations. The results reported by Ostendorf et al suggest that the presence of an inflammatory "pannus" may play a role in the development of bone alterations, at least at MR imaging. In the current study, all patients with SLE initially had inflammatory polyarthralgia with no signs of deforming arthropathy at clinical examination.
Cimmino et al (24), in describing two cases of mixed connective tissue disease, reported bone edema in the capitate bone that disappeared after treatment in one patient; such edema precedes the development of frank erosions in RA (2). Ejbjerg et al (26) recently studied the wrist and MCP joints of 28 healthy control subjects with contrast-enhanced MR imaging. MR images were evaluated according to the OMERACT RA-MRI scoring system. These authors occasionally found changes resembling mild synovitis or small bone erosions in the MCP and wrist joints, but signs of bone marrow edema were rare or absent and may be more specific for early RA. In our study, bone edema was not specific for early RA, although it was more common in patients with RA than in those without RA.
In addition to the presence of bone alterations in patients with SLE, the location of these bone alterations at the wrist and the MCP joints in the present study was similar to previously reported MR imaging findings in patients with RA (2,4,27). Our results are consistent with those of Ostendorf et al (11), who observed erosions at the MCP joints (primarily the second and third joints) and at the wrist joints in patients with SLE. The precise location of bone lesions within the carpus, however, was not determined in the study of Ostendorf et al. We did not see many signs of bone erosion in patients with primary Sjögren syndrome, especially at the MCP joints—unlike in patients with SLE. However, only limited conclusions can be drawn from a sample limited to five patients.
In patients with RA, edematous and inflammatory changes are confined within the joint capsule. In other conditions, such MR imaging abnormalities have been described far beyond the joint cavity, having been found to involve the periarticular soft tissues in patients with seronegative spondyloarthritis (20) and polymyalgia rheumatica (23). Jevtic et al (20) noted inflammatory changes involving the periarticular structures of the hand (ie, the collateral ligaments and soft tissues) in all of the patients with Reiter syndrome and in half of the patients with psoriatic arthritis that they evaluated. Similarly, McGonagle et al (23) noted extensive extracapsular edema in the soft tissues of the shoulder girdle in 64% of patients with polymyalgia rheumatica. More recently, in one of two patients with mixed connective tissue disease, Cimmino et al (24) noted pericapsular edema in a proximal interphalangeal joint. However, we did not find any abnormalities in the soft tissues adjacent to joint structures in patients with SLE or primary Sjögren syndrome, so we conclude that such abnormalities are not useful for distinguishing patients with early RA from those with SLE or primary Sjögren syndrome.
Our study had limitations. First, only patients with inflammatory polyarthralgia and no radiographic evidence of RA were selected to undergo MR imaging of both hands. However, in the everyday clinical management of early RA, many clinicians encounter patients who have only inflammatory polyarthralgia, are seronegative for rheumatoid factor, and exhibit no evidence of bone erosion on radiographs. Second, the sample size was small in each group (ie, patients with and patients without RA). MR imaging findings in the hands of patients with SLE or primary Sjögren syndrome should be evaluated in a larger study. Third, no true control group was evaluated.
In conclusion, the results of our study revealed that, in patients with inflammatory polyarthralgia of the hands and no evidence of erosive changes on radiographs, it may be impossible to differentiate patients with early RA from those with SLE or primary Sjögren syndrome by means of MR imaging.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Abbreviations: FLASH = fast low-angle shot • MCP = metacarpophalangeal • RA = rheumatoid arthritis • SE = spin echo • SLE = systemic lupus erythematosus • STIR = short inversion time inversion recovery
Authors stated no financial relationship to disclose.
Author contributions: Guarantors of integrity of entire study, N.B., A.C.; study concepts and design, all authors; literature research, N.B., A.C.; clinical studies, E.H., R.M.F.; data acquisition, N.B., E.H., R.M.F.; data analysis/interpretation, N.B., B.C., A.C.; statistical analysis, B.C.; manuscript preparation and editing, N.B.; manuscript revision/review, N.B., A.C.; manuscript definition of intellectual content and final version approval, all authors
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. J. Hodgson, P. O'Connor, and R. Moots MRI of rheumatoid arthritis image quantitation for the assessment of disease activity, progression and response to therapy Rheumatology, January 1, 2008; 47(1): 13 - 21. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Boutry, M. Morel, R.-M. Flipo, X. Demondion, and A. Cotten Early Rheumatoid Arthritis: A Review of MRI and Sonographic Findings Am. J. Roentgenol., December 1, 2007; 189(6): 1502 - 1509. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. G. Conaghan, P. O'Connor, a. P. Emery, N. Boutry, and A. Cotten MR Imaging Findings in the Hands Radiology, October 1, 2006; 241(1): 320 - 321. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
