Low-Field Compact Magnetic Resonance Imaging System for the Hand and Wrist in Rheumatoid Arthritis

Transcription

1 JOURNAL OF MAGNETIC RESONANCE IMAGING 23: (2006) Original Research Low-Field Compact Magnetic Resonance Imaging System for the Hand and Wrist in Rheumatoid Arthritis Hiroshi Yoshioka, MD, 1 * Satoshi Ito, MD, 2 Shinya Handa, BE, 3 Sadanori Tomiha, ME, 3 Katsumi Kose, PhD, 3 Tomoyuki Haishi, PhD, 4 Akito Tsutsumi, MD, 2 and Takayuki Sumida, MD 2 Purpose: To investigate the feasibility of an originally developed compact MRI system for evaluating rheumatoid arthritis (RA), and determine its advantages and disadvantages as an imaging modality for evaluating RA. Materials and Methods: We prospectively studied 13 healthy controls with no clinical symptoms of arthritis, and 13 patients with hand and wrist pains (including pain from RA) with a 0.2 T permanent-magnet compact MR imager. All MR images were obtained while the subjects were in a sitting position. Coronal three-dimensional spin-echo T1-weighted images and coronal two-dimensional short tau inversion recovery (STIR) images were obtained with image matrix and field of view (FOV) cm. Plain radiograph findings and MRI findings of patients were compared. Results: In three of the patients with suspected early RA (N 7), early RA was evaluated based on STIR images. All RA patients showed morphologic or signal intensity changes that allowed an evaluation of RA from MR findings. Four of five RA patients showed high signal intensity on STIR images in the wrist, proximal interphalangeal (PIP) joint, or metacarpophalangeal (MCP) joint that suggested synovitis. Multiple erosions in the hand and wrist were seen in four RA patients, with low signal intensity on T1-weighted images. Conclusion: RA was correctly evaluated, and early RA could be identified with the compact MRI system. However, the current system has limitations, such as the nonselective STIR sequence used and magnetic field inhomogeneity. Key Words: MR imaging; low-field MRI; compact MRI; rheumatoid arthritis; hand and wrist J. Magn. Reson. Imaging 2006;23: Wiley-Liss, Inc. MAGNETIC RESONANCE IMAGING (MRI), with its high soft-tissue contrast and absence of ionizing radiation, 1 Department of Radiology, Brigham and Women s Hospital, Boston, Massachusetts, USA. 2 Department of Rheumatology, University of Tsukuba, Tsukuba, Japan. 3 Institute of Applied Physics, University of Tsukuba, Tsukuba, Japan. 4 MR Technology Inc., Tsukuba, Japan. *Address reprint requests to: H.Y., Department of Radiology, Brigham and Women s Hospital, 75 Francis Street, Boston, MA hyoshioka@partners.org Received March 31, 2005; Accepted November 19, DOI /jmri Published online 2 February 2006 in Wiley InterScience (www. interscience.wiley.com). is a useful imaging tool for diagnosing musculoskeletal pathologies, including rheumatoid arthritis (RA). Numerous studies of RA have shown that MRI is much more sensitive than radiography or ultrasonography, especially for detecting early-stage RA (1 8). MRI is seven- to ninefold more sensitive than plain radiography for detecting erosions in early disease, and is able to detect lesions 6 12 months before they appear on plain radiographs (3,5,9 11). In one study, MRI identified more than twice as many erosions than ultrasonography and radiography, and was more sensitive than ultrasonography for detecting synovial disease (12). However, conventional whole-body, high-field MRI is more expensive (in terms of both startup costs and maintenance fees) and inconvenient for patients than radiography or ultrasonography. In addition, some metal objects, such as pacemakers and aneurysm clips, are contraindicated for whole-body MRI, and claustrophobic patients are sometimes unable to complete the examination. Low-field dedicated-extremity MR machines are now commercially available and have been applied to the evaluation of RA. Taouli et al (13) reported that conventional high-field MRI and low-field dedicated-extremity MRI were equivalent for evaluating RA synovitis. Using low-field dedicated-extremity MRI, Savnik et al (14) achieved a diagnostic accuracy for synovitis, bone marrow edema, joint effusion, and bone erosion of RA comparable to that of a high-field MRI system. Crues et al (15) reported that a portable MR system (0.2T) showed superior sensitivity for detecting bone damage compared to radiographs of the wrists and metacarpophalangeal (MCP) joints. They reported that MRI identified bony erosion in 95% of patients, compared to 59% by radiographs. Peterfy (8) remarked that the introduction of effective therapies for RA has increased the technical requirements for imaging in rheumatology, and that low-field extremity MRI offers adequate performance at a lower cost and with greater comfort and convenience for the patient. However, one disadvantage of these low-field MRI systems is that the field of view (FOV) is too small to enable an assessment of the hand and wrist in one examination or one sequence. Kose et al (16) recently developed a compact MRI system for measuring bone density of the calcaneus, and demonstrated its potential for that purpose. Their 2006 Wiley-Liss, Inc. 370

2 Compact MRI for Hand and Wrist in RA 371 Figure 1. (a) Overview of the compact MRI systems for the hand and wrist, and (b) hand positioning at MR examination. Each subject sat in front of the magnet with his or her hand inserted into the RF coil for acquisition of MR images. entire system consisted of a permanent magnet, gradient coil assembly, RF probe, and portable MR console. It was small and could be installed in a space of 2 m 2 m. Kurimoto et al (17) developed two compact MRI systems for the hand, and reported that these systems successfully detected several hand injuries within a reasonable imaging time. These results suggest that compact MRI has the potential to be useful for imaging the musculoskeletal region. Accordingly, our purpose in this study was to investigate the feasibility of using our originally developed compact MRI with a relatively large FOV for evaluating RA with the subject in a sitting position, and to determine the advantages and disadvantages of this compact MRI system as an imaging modality for evaluating RA. MATERIALS AND METHODS We prospectively studied 13 normal controls with no clinical symptom of arthritis, and 13 patients with hand and wrist pains, with an originally developed compact MR imager (Fig. 1). The normal control group included eight men and five women (22 48 years old, mean age 34.1 years). The patient group consisted of 13 women (29 68 years old, mean age 50.2 years). Table 1 presents their clinical profiles. Before the MR examination was performed, seven patients were suspected of having early RA based on arthralgia and swelling of the hand joints, an inflammatory reaction, positivity for rheumatoid factor, morning stiffness, and duration of symptoms, but had not yet met the 1987 revised American Rheumatology Association (ARA) criteria for the classification of RA. These patients were defined as patients with suspected early RA. One of the seven patients with suspected early RA was later diagnosed with ovarian cancer, paraneoplastic syndrome, and amyopathic dermatomyositis, and one showed soft-tissue swelling of the hand and fever of unknown origin. Five patients already met the revised ARA criteria. These patients were defined as RA patients. Therefore, the patient group consisted of subjects with suspected early RA, RA, and soft-tissue swelling. The study was approved by the institutional review board of the institute in which the MR system was placed. Each control subject and patient gave written informed consent. The compact MRI system consisted of a permanent magnet, gradient coil set, RF coil, and MRI console. The total installation space was 6m 2 (Fig. 1a). The magnet was placed in an electromagnetic shield room (1.6 m (W) 2.0 m (H) 2.4 m (D)) to prevent external noise. The specifications of the permanent magnet were as follows: magnetic field 0.21 T; gap 25 cm; homogeneity 25 ppm over 15 cm diameter spherical volume (dsv); and weight 1350 kg. For uniform RF ex- Table 1 Patient Profile Patient # Sex Age Clinical diagnosis Duration of symptom (years) 1 F 48 Early RA suspect 1 2 F 36 Soft tissue swelling, fever of unknown origin 1 3 F 68 Early RA suspect a 1 4 F 56 Early RA suspect 1 5 F 57 Early RA suspect 1 6 F 29 Early RA suspect 1 7 F 55 Early RA suspect 1 8 F 61 Early RA suspect 1 9 F 29 RA 3 10 F 70 RA F 51 RA 3 12 F 39 RA F 54 RA 12 a The patient was later diagnosed with ovarian cancer, paraneoplastic syndrome, and amyopathic dermatomyositis.

3 372 Yoshioka et al. Table 2 Comparison Between Plain Radiography Findings and MRI Findings Patient # Plain radiography findings MR findings 1 Diffuse periarticular osteoporosis Small cystic change at left distal metacarpal head, otherwise normal 2 No abnormality T1 low, T2 high-intensity soft tissue mass in left dorsal hand 3 Diffuse periarticular osteoporosis, right MCP joints erosion Right 2nd and 3rd MCP joint synovitis suspect 4 Mild periarticular osteoporosis, bilateral 5th MCP erosion Bilateral wrist synovitis early RA suspect 5 Mild periarticular osteoporosis Ganglion cyst on left 4th flexor tendon 6 Left 2nd PIP joint soft tissue swelling suspect Left 2nd PIP joint synovitis early RA 7 No abnormality Bilateral 1st and left 5th flexor tenosynovitis early RA 8 Bilateral 2nd and 3rd DIP joint OA change Small cystic change at right distal metacarpal head, otherwise normal 9 a Periarticular osteoporosis Bilateral wrist synovitis 10 a Advanced RA change Bilateral erosions, marked joint deformity and ankylosis of hands and wrists 11 a Bilateral scaphoid erosion, left 2nd MCP joint erosion Multiple erosions in bilateral wrists and, right 5th and left 2nd MCP joint 12 a Advanced RA change Multiple erosions in bilateral wrists, bilateral MCP joints, right 1st IP, and right 2nd PIP joint 13 a Advanced RA change Multiple erosions in bilateral wrists and carpal ankylosis a The patient met the criteria for diagnosis of RA. citation over a hand, we developed a solenoid RF coil with an oval aperture (8.5 cm (H) 21 cm (W)). The subject sat in front of the magnet and inserted his or her hand into the RF coil to generate MR images (Fig. 1b). Coronal three-dimensional (3D) spin-echo (SE) T1- weighted images (repetition time (TR)/echo time (TE) 200/16 msec for control group and 160/16 msec for patients) were obtained with an image matrix of , FOV of cm cm 6.4 cm, and one acquisition. The scan time was 6.8 min for the control group and 5.5 min for the patients. Coronal two-dimensional (2D) short tau inversion recovery (STIR) images (TR/TE/inversion time (TI) 1400/50/ 130 msec) were also obtained with an image matrix of , no slice selection, an FOV of cm cm, and two acquisitions. The scan time was 6 min. The scan time with TR/TE/TI 1000/50/120 msec was shorter for one patient because of pain (scan time 4.3 min). Both hands of each subject were scanned, and the total examination time, including patient positioning, was 40 min. For all patients, plain radiographs of both hands were obtained prior to the MR examination and reported by radiologists in our institute as a routine reading. No normal control groups underwent plain radiographic examination. The MR images were read by a radiologist (H.Y.) who was unaware of the radiographic findings. RESULTS All but one of the 26 subjects underwent successful MR examinations of both hand and wrist without complications. The one exception was patient 11, who showed a motion artifact because of pain. Only one of the control subjects showed an abnormal finding (a cystic change in the proximal first metacarpal bone without any clinical symptoms). The findings from the plain radiograph and MR images of the patients are shown in Table 2. In the patients with suspected early RA, three had an evaluation by MRI of early RA due to proximal interphalangeal (PIP) synovitis, wrist synovitis, and flexor tenosynovitis. The PIP synovitis showed soft-tissue swelling around the joint and high signal intensity in and around the joint on STIR images (Fig. 2). The evaluation of wrist synovitis was based on findings of high signal intensity on STIR images in the intercarpal and radiocarpal joints. Linear high signal intensity over the wrist and digits with the STIR sequence affirmed an evaluation of tenosynovitis (Fig. 3). All RA patients showed morphologic changes, including dislocation or ankylosis, or signal-intensity changes that permitted an evaluation of RA from MR findings. Four of five RA patients showed a high signal intensity in the wrist, MCP, or PIP joint that suggested synovitis on STIR images. One patient (patient 10) showed no high signal intensity on STIR images, but had marked bilateral ulnar deviated deformity of the hand and wrist ankylosis. Multiple erosions in the hand and wrist were seen in four RA patients who exhibited low signal intensity on T1-weighted images (Fig. 4). DISCUSSION MRI has become an important imaging modality for evaluating RA. Bony erosion is an essential finding in the evaluation and assessment of RA, and MRI is highly sensitive for detecting erosions in the hands and wrists of patients with RA (3,5,10,15,18). Many MR studies have emphasized the usefulness of gadolinium contrast

4 Compact MRI for Hand and Wrist in RA 373 Figure 2. MRI of the left hand of a 29-year-old woman with suspected early RA (patient 6). Her second PIP joint was swollen and showed slightly low signal intensity on (a) the T1- weighted image, and high signal intensity on (b) the STIR image (arrow), suggesting early RA. material in assessing active synovitis and destructive pannus (3,19 22). Most of these MR examinations were performed with high-field MRI systems. These systems can provide images with higher signal-to-noise ratios (SNRs) and higher spatial resolution than low-field MRI systems. However, conventional whole-body, high-field MRI is expensive and inconvenient for patients and has some contraindications, such as the presence of metal objects (e.g., pacemakers, aneurysm clips, and cochlear implants) in the subject and claustrophobia. In general, the hand and wrist joints are among the first to be affected in RA, and they are of particular interest in the assessment of patients with suspected early RA (13,23). Therefore, for screening and early treatment of RA (especially early-stage RA), dedicated MRI of the hand may be useful for assessing RA if the system is less expensive, is more easily installed, improves patient comfort, and entails less patient risk than whole-body high-field MRI. The design of the low-field compact MRI system for the hand and wrist in this study was originally based on these concepts. Taouli et al (13) and Savnik et al (14) recently compared low-field dedicated-extremity MRI with high-field MRI for evaluating RA of the hand and wrist. Taouli et al (13) reported that the two systems performed equally well for cross-sectional grading of bone erosions, joint- Figure 3. T1-weighted images (a: left; b: right) and STIR images (c: left; d: right) of both hands of a 55-year-old woman with suspected early RA (patient 7). Linear high signal intensities over the bilateral wrists and first finger, and the left fifth finger on STIR (arrows) suggest tenosynovitis.

5 374 Yoshioka et al. Figure 4. MRI of both hands of a 39-year-old woman with RA (patient 12). a and b: T1- weighted images showed multiple low signal intensity erosions of the bilateral interphalangeal joints of the thumb, MCP joints, and wrist joints. c and d: STIR images show synovitis of the above joints as high signal intensity. space narrowing, and synovitis in RA. Both MRI techniques detected approximately twice as many erosions as detected by radiography (13). The Savnik et al (14) study showed that the volume of synovial membrane determined by extremity MRI was significantly correlated with and not significantly different from that determined with high-field MRI with gadolinium injection. Intraobserver variation was high for the two MR systems. Therefore, the performance of low-field dedicatedextremity MRI would be the same as that of high-field MRI for the evaluation of RA, including early-stage RA. The use of dedicated-extremity MRI for the evaluation of RA has several advantages over whole-body MRI. Compared to a whole-body MRI system, dedicated-extremity MRI requires a smaller installation space, costs less, offers greater patient comfort, and eliminates the problem of claustrophobia and potential biohazards associated with the presence of metal in or on the patient, since only the limb of interest is placed into the magnet bore. Therefore, it may be more feasible than conventional high-field MRI for evaluating patients with arthritis of the small joints (13). Savnik et al (14) reported that 64% of patients with arthritis of the hand and wrist preferred 0.2T extremity MRI to 1.5T high-field MRI because the former was more comfortable, less claustrophobic, and quieter. Of their patients, 20% reported experiencing mild to severe claustrophobia during the high-field MR examination. Lindegaard et al (10) also reported advantages of extremity MRI over whole-body MRI with regard to cost, comfort, and claustrophobia. In our study, these three Cs (cost, comfort, and claustrophobia), along with the small installation space, no maintenance, and no ionizing radiation, were the greatest advantages of extremity MRI. In addition, it had the technical advantage of a larger FOV (20.48 cm cm), which allowed us to obtain MR images that included the wrist joint to the PIP joint. Some earlier studies of extremity MRI had to use a smaller FOV and thus had limited anatomic coverage (13 15). With a smaller FOV, fewer joints can be examined in a given imaging series. In the present study, examinations of both hands, including positioning of subjects, could be performed within 40 min. Low-field compact MRI is insensitive to metal objects implanted anywhere but in the hand and wrist, and thus reduces the risk associated with the presence of metal in the body. Also, as Crues et al (15) mentioned, the unique features of the portable MR system allow it to be readily used in an office or point-of-care setting (e.g., a rheumatology office or clinic), improving overall patient management and operational efficiency. Generally, however, low-field dedicated-extremity MRI has the limitations of lower SNRs and longer acquisition times compared to conventional high-field MRI (13). Moreover, the present study had several limitations, as described below. First, the STIR images were obtained with a nonslice-selective sequence because STIR with multiple

6 Compact MRI for Hand and Wrist in RA 375 slices requires a longer scan time with an SE sequence. Thus, it was not possible to look at detailed anatomy with STIR. Also, it may be difficult to point out bone marrow edema. Non-slice-selective images were inadequate for assessing a large soft-tissue mass (in patient 2), since they could not show abnormal findings in detail because of the signal intensity of the normal structure overlying the lesion. However, joint effusion, synovitis, and tenosynovitis with STIR produced bright signals and enabled us to evaluate RA. Although we were able to evaluate erosion in 3D T1-weighted images in this study, we plan to use a multislice STIR sequence within a reasonable scan time as the next development in our compact MRI system. Second, we could not use a spectral fat-suppression technique because of close spectral peaks of fat and water in this study. On a low-field MR machine, it is difficult to obtain fat-suppressed T1-weighted images using a spectral fat-suppression technique, which is often combined with the use of gadolinium contrast. Fat-suppressed T1-weighted images with gadolinium were shown to be useful for diagnosing early and active RA in many earlier studies (4 6,21). In one study (13), synovitis was visible on coronal T1-weighted 3D gradient-recalled echo (GRE) and STIR low-field MR images as a fluid signal distending the joint cavity. Synovial tissue cannot be reliably differentiated from synovial fluid with these sequences. Discriminating between these two components of the synovial cavity typically requires the use of IV contrast materials. However, this increases the cost and time required for the examination. It would be difficult to study both hands and wrists, such as performed in this study, if contrast material were used. For screening both hands for RA, non-enhanced water-sensitive sequences, such as STIR, may be desirable. Third, in this study there was no established gold standard with which to compare the MRI and radiograph results, because no pathology specimens were available from the patients. It has not been proven that MR erosions are true erosions or that MR erosions will develop into radiographic erosions (1). Nevertheless, three patients with suspected early RA started antiinflammatory treatment after MRI evaluation, and the treatment was effective (i.e., joint pain and morning stiffness improved) in all three. Therefore, MRI helped physicians make decisions about treatment in a timely fashion. Fourth, image artifacts caused by inhomogeneity of the static field were sometimes seen around the wrists. To resolve this artifact, we should have shifted the hands slightly farther into the magnet. Fifth, our system cannot provide spatial resolution as high as that achievable with high-field MRI. However, as Peterfy (8) commented, the needs of radiology are different from those of mainstream rheumatology, and rheumatologists do not need highly sophisticated pulse sequence and contrast mechanisms. The ability to detect occult bone erosions and synovitis would be adequate to allow rheumatologists to initiate patient management and treatment. Sixth, typical erosions in RA measure 2 3 mm in maximum diameter. Our protocol uses 4-mm-thick images with mm in-plane resolution. With this resolution, changes in erosion that occur over small periods of time will not be reliable. We may combine a small surface coil with this system to detect a subtle change on high-resolution images in the future. Finally, the number of subjects in this study was small. We plan to perform a study with more patients after we have improved our compact MRI system, including the pulse sequence, magnet, gradient coils, and RF coil. To ensure the reliability of our compact MRI system, we need to assess intra- and interreader reproducibility for the evaluation of RA in future studies. In conclusion, normal controls and patients with hand and wrist pain, including those with suspected early RA and RA that met RA criteria, were examined with a prototype low-field hand and wrist dedicated compact MRI system. This system was able to evaluate RA correctly and suggest early RA. However, the system still has limitations, such as the use of a non-selective STIR sequence and magnetic field inhomogeneity. A large number of cohort studies are still necessary to improve the system. REFERENCES 1. Klarlund M, Ostergaard M, Gideon P, Sorensen K, Jensen KE, Lorenzen I. Wrist and finger joint MR imaging in rheumatoid arthritis. Acta Radiol 1999;40: McQueen FM. Magnetic resonance imaging in early inflammatory arthritis: what is its role? Rheumatology (Oxf) 2000;39: Klarlund M, Ostergaard M, Jensen KE, Madsen JL, Skjodt H, Lorenzen I. Magnetic resonance imaging, radiography, and scintigraphy of the finger joints: one year follow up of patients with early arthritis. The TIRA Group. Ann Rheum Dis 2000;59: Ostergaard M, Szkudlarek M. Imaging in rheumatoid arthritis why MRI and ultrasonography can no longer be ignored. Scand J Rheumatol 2003;32: McQueen FM, Stewart N, Crabbe J, et al. Magnetic resonance imaging of the wrist in early rheumatoid arthritis reveals a high prevalence of erosions at four months after symptom onset. Ann Rheum Dis 1998;57: Sugimoto H, Takeda A, Masuyama J, Furuse M. Early-stage rheumatoid arthritis: diagnostic accuracy of MR imaging. Radiology 1996;198: Sugimoto H, Takeda A, Hyodoh K. Early-stage rheumatoid arthritis: prospective study of the effectiveness of MR imaging for diagnosis. Radiology 2000;216: Peterfy CG. Is there a role for extremity magnetic resonance imaging in routine clinical management of rheumatoid arthritis? J Rheumatol 2004;31: McQueen F, Lassere M, Edmonds J, et al. OMERACT rheumatoid arthritis magnetic resonance imaging studies. Summary of OMER- ACT 6 MR imaging module. J Rheumatol 2003;30: Lindegaard H, Vallo J, Horslev-Petersen K, Junker P, Ostergaard M. Low field dedicated magnetic resonance imaging in untreated rheumatoid arthritis of recent onset. Ann Rheum Dis 2001;60: McQueen FM, Benton N, Crabbe J, et al. What is the fate of erosions in early rheumatoid arthritis? Tracking individual lesions using x rays and magnetic resonance imaging over the first two years of disease. Ann Rheum Dis 2001;60: Hoving JL, Buchbinder R, Hall S, et al. A comparison of magnetic resonance imaging, sonography, and radiography of the hand in patients with early rheumatoid arthritis. J Rheumatol 2004;31: Taouli B, Zaim S, Peterfy CG, et al. Rheumatoid arthritis of the hand and wrist: comparison of three imaging techniques. AJR Am J Roentgenol 2004;182: Savnik A, Malmskov H, Thomsen HS, et al. MRI of the arthritic small joints: comparison of extremity MRI (0.2 T) vs high-field MRI (1.5 T). Eur Radiol 2001;11:

7 376 Yoshioka et al. 15. Crues JV, Shellock FG, Dardashti S, James TW, Troum OM. Identification of wrist and metacarpophalangeal joint erosions using a portable magnetic resonance imaging system compared to conventional radiographs. J Rheumatol 2004;31: Kose K, Matsuda Y, Yoshioka H, et al. Development of a compact MRI system for trabecular bone volume fraction measurements. Magn Reson Med 2004;52: Kurimoto T, Furuya T, Kose K, et al. Compact hand MRI systems using permanent magnets. In: Proceedings of the 12th Annual Meeting of ISMRM, Kyoto, Japan, p McGonagle D, Conaghan PG, O Connor P, et al. The relationship between synovitis and bone changes in early untreated rheumatoid arthritis: a controlled magnetic resonance imaging study. Arthritis Rheum 1999;42: Jevtic V, Watt I, Rozman B, et al. Precontrast and postcontrast (Gd-DTPA) magnetic resonance imaging of hand joints in patients with rheumatoid arthritis. Clin Radiol 1993;48: Konig H, Sieper J, Wolf KJ. Rheumatoid arthritis: evaluation of hypervascular and fibrous pannus with dynamic MR imaging enhanced with Gd-DTPA. Radiology 1990;176: Nakahara N, Uetani M, Hayashi K, Kawahara Y, Matsumoto T, Oda J. Gadolinium-enhanced MR imaging of the wrist in rheumatoid arthritis: value of fat suppression pulse sequences. Skeletal Radiol 1996;25: Jevtic V, Watt I, Rozman B, et al. Prognostic value of contrast enhanced Gd-DTPA MRI for development of bone erosive changes in rheumatoid arthritis. Br J Rheumatol 1996;35: Scott DL, Coulton BL, Popert AJ. Long term progression of joint damage in rheumatoid arthritis. Ann Rheum Dis 1986;45: