QUANTITATIVE ASSESSMENT OF THE SYNOVIAL MEMBRANE IN THE RHEUMATOID WRIST: AN EASILY OBTAINED MRI SCORE REFLECTS THE SYNOVIAL VOLUME

Transcription

1 British Journal of Rheumatology 1996;35: QUANTITATIVE ASSESSMENT OF THE SYNOVIAL MEMBRANE IN THE RHEUMATOID WRIST: AN EASILY OBTAINED MRI SCORE REFLECTS THE SYNOVIAL VOLUME M. 0STERGAARD,*-t M. HANSEN,*t M. STOLTENBERG*t and I. LORENZEN* 'Department of Rheumatology and ^Danish Research Centre of Magnetic Resonance, Hvidovre Hospital and %Department of Rheumatology, Herlev Hospital, University of Copenhagen, Denmark SUMMARY Determination of the synovial membrane volume in the rheumatoid arthritis (RA) wrist by gadolinium-dtpa-enhanced MRI is introduced. Moreover, dynamic imaging and an MRI score of synovial hypertrophy, based on gradings in six regions, are evaluated as substitutes of the time-consuming volume calculations. Twenty-six RA wrists were examined. Synovial membrane volumes ranged from 1 to 20 ml (median 9 ml). Synovial hypertrophy scores were highly correlated to synovial volumes (Spearman r = 0.88; P < 10~* for uncorrelated values). The volumes and scores were significantly higher in wrists with joint swelling and/or joint tenderness than in wrists without these signs (Mann-Whitney, both P < 0.05). Suboptimal slice selection made dynamic imaging uninformative. MRI allows quantification of the synovial volume in the rheumatoid wrist. The volume is related to clinical signs of inflammation, but may also give information about the cumulated synovial proliferation in the joint. An easily obtained score of synovial hypertrophy reflects the synovial volume and may thus be a useful marker of synovial involvement. KEY WORDS: Rheumatoid arthritis, Synovitis, Arthritis, Magnetic resonance imaging, Nuclear magnetic resonance, Gadolinium, Gadopentetate dimeglumine, Wrist. THE synovial membrane is the primary site of rheumatoid inflammation. In rheumatoid arthritis (RA), as in other inflammatory arthritides, the mass of synovial tissue is markedly increased [1,2]. Thus, a quantitative measure of the amount of synovial membrane may provide useful information on disease severity and/or activity. Magnetic resonance imaging (MRI) allows direct visualization of the inflamed synovial membrane in knees [3-6] and wrists [4,7-9]. Determination of the synovial membrane volume of the knee is possible by MRI, after i.v. injection of the contrast agent gadolinium-dtpa (Gd-DTPA) [10-13]. In these knee joint studies, the method was found to be reproducible [10, 13] and reasonably accurate (the maximal analytical error on volume determinations was ~20% [13]). The earliest changes in RA are often found in the wrists, hands and feet [14, 15]. Furthermore, destructive changes in the hands and wrists are considered the best radiological indicator of overall joint damage [15]. Consequently, application of the methods on the wrist and other smaller joints is desirable. The present study introduces estimation of the synovial membrane volume in the rheumatoid wrist, achieved by computer-assisted outlining of the synovial membrane on post-gd-dtpa MR images. Since the procedure of volume determination is quite time consuming, we evaluated whether a semi-quantitative scoring method of synovial hypertrophy in the wrist, introduced in a previous study [16], or dynamic Submitted S December 1995;revisedversion accepted 3 April Correspondence to: M. 0stergaard, Department of Rheumatology, Hvidovre Hospital, DK-2650 Hvidovre, Denmark. Gd-DTPA-enhanced MRI [3,4,17-19], may replace the volume calculations. PATIENTS AND METHODS MRI of the wrist was performed in 26 patients with RA, fulfilling the ARA 1987 classification criteria of RA [20]. All patients had active disease, defined as arthritis in at least three joint areas [right or left proximal interphalangeal (PIP), metacarpophalangeal (MCP), wrist, elbow, knee, ankle or metatarsophalangeal (MTP) joint] with joint tenderness and/or joint swelling and two out of the three following criteria: morning stiffness ^ 1 h, erythrocyte sedimentation rate (ESR) > 35 mm/h and serum C-reactive protein (s-crp) > 150 nmol/1. The median age of the patients was 60 yr (range yr), while the median duration of disease was 3 yr (range 3 months-22 yr). MRI The MR images were obtained using a 1.5 Tesla magnetom unit (Siemens, Erlangen, Germany), equipped with a transmit-receive knee coil. The patient was positioned on the side with the opposite hand in front of the head, in the knee coil. The right wrist was examined in 24 patients. In two patients, severe pain in the left shoulder made imaging of the right wrist impossible. In these two patients, the left wrist was examined. Continuous coronal and transversal Tlweighted spin-echo MR images (TR/TE/slice thickness = ms/15-17 ms/3 mm) were obtained. A series of 30 Tl-weighted FLASH (fast low angle shot) images (TR/TE/flip angle/slice thickness = 40 ms/ 12 ms/70 /5 mm), each with an imaging time of 10 s, was then performed in the same, pre-selected, transversal slice (dynamic imaging). During the second British Society for Rheumatology

2 966 BRITISH JOURNAL OF RHEUMATOLOGY VOL. 35 NO. 10 Synovial membrane volume By means of the image-processing software package XPrime, installed on a Sun Sparc 10 computer (Unix), the synovial membrane of each transversal slice was outlined and the areas automatically calculated. The outlining was done on post-gd-dtpa images, displayed on a computer screen, by means of a computer mouse. The outlining was guided by subtraction images. The total volume of synovial membrane (Vol.^) was calculated by summation of the slices using the following formula: Vol^ = HAr^ x ST) where ST is the slice thickness and Ar^j represents the area of synovial membrane in slice /. Grading and scoring of synovial membrane hypertrophy An MRI grading of synovial hypertrophy in each of six parts of the wrist was performed on the pre- and post-contrast coronal images: 0: no visible enhance- FIG. ment; 1: linear enhancement (width <2 mm); 2: band-like enhancement (width >2mm, but <4mm); 3: diffuse enhancement (width 3*4 mm). Examples are given in Fig. 1. An MRI score (0-18) of synovial membrane hypertrophy was calculated by adding the scores from the six regions. The six regions, attempted to be representative of the wrist, were as follows: (a) the distal radioulnar joint; (b) the ulnar part of the radiocarpal joint (including around the ulnar styloid); (c) the radial part of the radiocarpal joint; (d) the intercarpal joints (defined as the S-shaped line of joints between the proximal and distal row of carpal bones); (e) the first carpometacarpal joint; (f) the 2nd-5th carpometacarpal joint. Early synovial enhancement (dynamic FLASH imaging) The image-processing software package XPrime allowed subtraction of images and outlining of areas of interest. The FLASH image of r0 was subtracted from the image of t-m (Fig. 2). On this subtraction image, the synovial membrane, that showed signal enhancement, was outlined. The mean signal intensity of the outlined area, i.e. the synovial membrane of the slice, to each time was automatically calculated by the computer. The relative synovial enhancement per second during the first 55 s (rate of early synovial enhancement; ) was calculated by the following formula: x 100% where Sh and SFa are the signal intensities before and 55 s after contrast injection, respectively. Evaluation of the enhancement after 55 s was chosen because a previous study showed maximal enhancement difference between knees with clinically active and clinically inactive arthritis in the interval from 50 to 90 s after Gd-DTPA injection [17]. l(a, b). period, 0.05 mmol Gd-DTPA/kg body weight was injected into a cubital vein, while the patient remained in the same position in the MR unit. The FLASH sequence covered the enhancement during the first 285 post-contrast seconds. The dynamic FLASH sequence only allowed us to examine one slice. We chose to pre-select the transversal slice through the hook of the hamate. This slice was chosen because transversal slices are least vulnerable to position differences and movement of the wrist, and because the slice would be easy to reproduce at subsequent examinations. Finally, the spin-echo sequences were repeated: the transversal images first (5-10 min after Gd-DTPA injection), then the coronal images (10-15 min after Gd-DTPA injection). In all sequences, the matrix size was x 256 and the field of view (FOV) mm.

3 0STERGAARD ET AL.: MRI-DETERMINED SYNOVIAL VOLUMES IN RA WRISTS 967 (c) Statistical methods Non-parametric methods were used to analyse the data. The Mann-Whitney test (two-sample rank sum test) was used to analyse differences between groups of patients. Analysis of statistical correlation was performed by the Spearman test of rank correlation. RESULTS Twenty wrists showed clinical signs of active synovitis of the examined wrist, i.e. joint swelling and/or joint tenderness, while these signs were absent in six wrists. On the Tl-weighted spin-echo MR images, the signal Fio. 1. Tl-weighted spin-echo images, (a-c) Transversal images obtained (a) before and (b) after i.v. injection of the contrast agent Gd-DTPA. After Gd-DTPA, the signal intensity of the synovial membrane (arrows) has increased markedly, (c) To illustrate the procedure of volume calculation, the synovial membrane has been outlined (white lines) on the post-gd-dtpa image. The computer automatically calculates the size of the outlined areas, and the synovial membrane volume is determined by summation of the areas of all the transversal slices, (d-e) Coronal images obtained (d) before and (e) after i.v. Gd-DTPA. Diffuse post-gd-dtpa synovial enhancement (synovial membrane hypertrophy grade 3; width ^ 4 mm) is seen in the ulnar part of the radiocarpal joint (straight black arrow). Band-like enhancement (Grade 2; width 5 2 mm, but < 4 mm) is seen in the 5th carpometacarpal joint and in the distal radioulnar joint (curved arrow). Linear enhancement (Grade 1, width < 2 mm) is found in the 1st carpometacarpal joint (white arrows), while no synovial enhancement (Grade 0) can be found in the intercarpal joints (open arrows). Bases of the 1st and 5th metacarpal bone (m), the trapezium (t), the hamate (h), the scaphoid (s) and the triquetrum (i).

4 968 BRITISH JOURNAL OF RHEUMATOLOGY VOL. 35 NO. 10 joint swelling and/or tenderness, the synovial volume ranged from 2 to 20 ml (median 11 ml). In wrists without these signs, the synovial volume ranged from 1 to 9 ml (median 4 ml). The synovial volume was statistically significantly higher in wrists with than in wrists without joint swelling and/or joint tenderness (Mann-Whitney, P < 0.05). The MRI score of synovial hypertrophy ranged from 2 to 16 (median 10) (Table I). Scores in wrists with/without clinical signs of synovial inflammation (b) Fio. 2. Transversal FLASH images through the hoolc of the hamate, (a-c) Same patient as in Fig. 1. FLASH images obtained (a) before (to) and (b) 285 s after (tns) Gd-DTPA injection. The signal intensity of the synovial membrane (arrows) has increased considerably from it to tni. (c) Subtraction image (b a; tm to). This image was used for outlining of the synovial membrane. Except for vessels (arrows), only the synovial membrane appears white. In this patient, the outlining was unproblematic. (d) The tm U> subtraction image of another patient Vessels (arrows) are the only structures which appear intensely white. In this patient, no synovial membrane could be identified in the pre-selected slice, thus REE*, could not be determined. intensity of the synovial membrane increased markedly following i.v. injection of Gd-DTPA. Consequently, it was possible to identify and outline the synovial membrane (Fig. 1). Enhancing vessels were, due to shape, appearance and localization, easily differentiated from the synovial membrane, and were not included in the outlined areas. The MRI-determined synovial membrane volume ranged from 1 to 20 ml (median 9 ml) (Table I). In wrists with clinical signs of synovial inflammation, i.e.

5 0STERGAARD ET AL.: MRI-DETERMINED SYNOVIAL VOLUMES IN RA WRISTS 969 Number of wrists Synovia] membrane volume Synovial hypertrophy score TABLE I Synovial membrane volumes and scores in clinically active and clinically inactive wrists Total 26 9 ml (1-20 ml) 10 (2-16) Joints with swelling and/or tenderness ml (2-20 ml) 11 (5-16) Joints without swelling and tenderness 6 4 ml (1-9 ml) 8.5 (2-12) Mann-Whitney test P < 0.05 P < 0.05 Median values are given, with the range in parentheses. The P values indicate statistically significant differences between the joints with and the joints without swelling and/or tenderness, at the mentioned level of significance. ranged from 5 to 16 (median 11) and from 2 to 12 (median 8.5), respectively. The difference was statistically significant (Mann-Whitney, P < 0.05). The rate of early synovial enhancement (REE^) was calculated from a series of FLASH MR images, obtained in the transversal slice through the hook of the hamate. In 12 wrists, including five of six wrists without clinical signs of inflammation, no synovial membrane could be differentiated with certainty in this pre-selected slice (Fig. 2). In the 14 wrists in which the synovial membrane was distinguishable, REE^n ranged from 0.4 to 2.2%/s (median 1.2%/s). Synovial volumes and scores were highly statistically correlated (Fig. 3). The Spearman correlation coefficient r was 0.88 (P < 10" 8 for uncorrelated values). The early synovial enhancement (REE) was not statistically correlated to the synovial scores (r = 0.43; P = 0.12) and volumes (r = 0.31; P = 0.28). Neither REE, the synovial score nor the synovial volume were correlated to ESR or s-crp. DISCUSSION The present study introduces estimation of the synovial membrane volume in the rheumatoid wrist, achieved by outlining of the synovial membrane on post-gd-dtpa MR images. The synovial volume was statistically significantly higher in wrists with clinical signs of synovitis than in clinically inactive wrists. This observation is in accordance with findings in arthritic knees [10, 12] O o On gross examination, the mass of the inflamed synovial membrane is dramatically increased. The macroscopic thickening of the synovium reflects vascular congestion, oedema and cellular infiltration, as well as synovial lining hyperplasia and pannus tissue formation [1,2]. Consequently, determination of the synovial volume would be expected to give information about disease activity and/or severity in RA. Following i.v. administration of the contrast agent Gd-DTPA, MRI allows visualization of the inflamed synovial membrane. Within clinically uniform groups, the synovial membrane volume varied considerably, e.g. in wrists with clinical signs of active synovitis, the synovial volume ranged from 2 to 20 ml. It is likely that this interval reflects clinically significant differences in disease activity/severity. However, longitudinal studies are needed to clarify the prognostic value of synovial volume measurements. The synovial membrane volume was not statistically correlated to ESR and s-crp. This is not surprising, since the laboratory parameters sum up the effects from all joints, while only one joint was examined by MRI. Furthermore, the synovial volume is probably not determined solely by the present inflammatory activity, but is also influenced by the cumulated synovial proliferative activity in the joint. Except for one recent study examining two wrists [21], synovial volume estimation by MRI has hitherto been restricted to knee joints [10-12], mainly due to technical limitations of MRI. Determination of the o o Synovial membrane volume (ml) Fio. 3. Synovial membrane volume versus synovial hypertrophy score. The linear regression line is shown. The non-parametric Spearman correlation coefficient is 0.88 (P < 10"' of uncorrelated values). 20

6 970 BRITISH JOURNAL OF RHEUMATOLOGY VOL. 35 NO. 10 synovial volume in smaller joints such as wrists and finger joints will be of greater clinical significance, since the earliest changes in RA are most often found in these joints [14, 15]. Quantitative assessment of the synovial volume even in finger joints will probably soon be possible, since dedicated wrist coils, MR units designed exclusively for limb examination and new software constantly improve image quality, image processing and clinical applicability. The main problem involved in delineating the synovial membrane is generally assumed to be the distinction from joint fluid [22, 23], because the signal intensity of the joint fluid also increases after Gd-DTPA injection, even though at a much slower rate than the synovium [23-25]. In a previous study, preand post-aspiration volumes were measured in knee joints [13]. No significant systematic misinterpretation of the borderline between joint fluid and synovium was found. Thus, in arthritic knee joints, the observed effusion-synovium borderline appears to be reliable at least within the initial 15 min after Gd-DTPA injection [6, 13]. In wrists, the anatomical structures are considerably smaller. Given the slice thickness of mm used in this study and in general [4, 7-9, 16, 21, 26, 27], it must be expected that partial volume artefacts (volume averaging effects) are of relatively greater importance than in knees. In the near future, thinner slices ( mm), obtained by three-dimensional gradient echo sequences and dedicated coils, will probably minimize this problem. Synovial volume estimations have hitherto been obtained either by manual computer-assisted outlining based on visual analysis of the images [10, 12, 13], as in the present study, or by semi-automatic computerized counting of pixels with a post-gd-dtpa signal enhancement above a certain threshold [11,21]. The semi-automatic methods are extremely sensitive to the enhancement thresholds chosen [11]. Manual methods appear more reliable, but are also more time demanding. The solution is probably to optimize the automated methods, and then use the manual methods as a reference, e.g. in order to select the correct enhancement threshold. Subsequently, the optimized automated methods may be used in clinical trials. Until fast and reliable methods for the determination of wrist synovial volumes have been developed, semi-quantitative measures of synovial involvement could be useful, for instance, in larger clinical trials. Several approaches have been published [9, 16, 26, 27]. In a previous study [16], a score of wrist synovial hypertrophy, based on coronal Tl -weighted spin-echo MRI, was introduced. The score was zero in healthy joints and related to the clinical inflammatory activity of the examined arthritic joints. In the present study, the score was furthermore highly statistically correlated with the MRI-determined synovial volume (Fig. 3), and it thus appears to be a quite adequate substitute for the presently very time-consuming volume measurements. Several studies on knee joints have strongly indicated that the early synovial post-gd-dtpa enhancement on Tl-weighted dynamic FLASH MR images is related to the inflammatory activity of the joint [3, 11, 17, 19]. In the present study, the rate of early synovial enhancement (REE^) was calculated from a transversal FLASH image through the hook of the hamate. This slice was chosen because transversal slices are least vulnerable to position differences and movement of the wrist, and because the slice would be easy to reproduce at subsequent examinations. Unfortunately, the amount of synovial membrane in this slice was generally small, making REE^, measurements less representative and less reliable. The synovial membrane could only be differentiated with certainty on the FLASH images of 14 out of 26 wrists. Thus, a comparison of the rate of early synovial enhancement in clinically active and clinically inactive wrists was not possible in this study. In future studies, a mid-joint coronal slice, which would be more representative of the wrist, must be chosen. The cost of new tests and procedures must always be considered prior to general use. MRI is still expensive, even though the cost is constantly decreasing. If future studies show that information of clinical and prognostic significance is provided by MRI, the cost may, however, easily be compensated for. In summary, MRI allows quantitative estimation of the synovial membrane volume in the rheumatoid wrist. The volume is related to the clinical inflammatory activity of the joint examined, but may also give information about the cumulated synovial proliferative activity in the joint. Longitudinal studies are needed to clarify the prognostic value of synovial volume measurements. An easily obtained score of synovial hypertrophy is highly statistically correlated with the synovial volume, and may thus be a useful marker of synovial mass, for instance in clinical trials, at least until fast and reliable automatic volume determination is possible. Suboptimal slice selection made dynamic imaging uninformative in this study. ACKNOWLEDGEMENTS We acknowledge the Foundation of , the University of Copenhagen, the Thomas & Elisabeth Fralund Nielsen Foundation, the Danish Rheumatism Association and the Danish Medical Research Council for financial support. Schering Diagnostika, Denmark, is acknowledged for providing the contrast agent. M0 would also like to thank Professor Ole Henriksen for valuable scientific guidance, and Jens Arnth Jensen and Poul Ring for developing the image-processing software package XPrime. REFERENCES 1. Resnick D. Common disorders of synovium-lined joints: pathogenesis, imaging abnormalities, and complications. Am J Roentgenol 1988;151: Fircstein G. Rheumatoid arthritis. Rheumatoid synovitis and pannus. In: Klippel JH, Dieppe PA, eds. Rheumatology. London: Mosby-Year Book Europe Ltd, 1994: Konig H, Sieper J, Wolf KJ. Rheumatoid arthritis:

7 0STERGAARD ET AL.: MRI-DETERMINED SYNOVIAL VOLUMES IN RA WRISTS 971 Evaluation of hypcrvascular and fibrous pannus with dynamic MR imaging enhanced with gd-dtpa. Radiology 1990; 176: Reiser MF, Bongartz GP, Erlemann R et al. Gadolinium-DTPA in rheumatoid arthritis and related diseases: First results with dynamic magnetic resonance imaging. Skel Radiol 1989;18: Adam G, Dammer M, Bohndorf K, Christoph R, Fenke F, Gunther RW. Rheumatoid arthritis of the knee: Value of gadopentetate dimeglumine-enhanced MR imaging. Am J Roentgenol 1991;156: Bjorkengren AG, Geborek P, Rydholm U, Holtas S, Petterson H. MR imaging of the knee in acute rheumatoid arthritis: Synovial uptake of gadolinium- DOTA. Am J Roentgenol 1990;155: Corvetta A, Giovagnoni A, Baldelli S et al. MR imaging of rheumatoid hand lesions: comparison with conventional radiology in 31 patients. Clin Exp Rheumatol 1992;10: Jorgensen C, Cyteval C, Anaya JM, Baron MP, Lamarque JL, Sany J. Sensitivity of magnetic resonance imaging of the wrist in very early rheumatoid arthritis. Clin Exp Rheumatol 1993;11: 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: stergaard M, Gideon P, Henriksen O, Lorenzen I. Synovial volume a marker of disease severity in rheumatoid arthritis? Quantification by MRI. Scand J Rheumatol 1994;23: Waterton JC, Rajanayagam V, Ross BD, Brown D, Whittemore A, Johnstone D. Magnetic resonance methods for measurement of disease progression in rheumatoid arthritis. Magn Reson Imag 1993;ll: stergaard M, Stoltenberg M, Gideon P et al. Effect of intraarticular osmic acid on synovial membrane volume and inflammation, determined by magnetic resonance imaging. Scand J Rheumatol 1995;24: stergaard M, Stoltenberg M, Henriksen O, Lorenzen I. The accuracy of MRI-determined synovial membrane and joint effusion volumes in arthritis. A comparison of pre- and post-aspiration volumes. Scand J Rheumatol 1995;24:305-ll. 14. Gordon DA, Hastings DE. Rheumatoid arthritis. Clinical features: Early, progressive and late disease. In: Klippel JH, Dieppe PA, eds. Rheumatology. London: Mosby-Year Book Europe Ltd, 1994: Scott DL, Coulton BL, Popert AJ. Long term progression of joint damage in rheumatoid arthritis. Ann Rheum Dis 1986;45: stergaard M, Gideon P, Serensen K et al. Scoring of synovial membrane hypertrophy and bone erosions by MR imaging in clinically active and inactive rheumatoid arthritis of the wrist. Scand J Rheumatol 1995;24: stergaard M, Lorenzen I, Henriksen O. Dynamic gadolinium-enhanced MR imaging in active and inactive immunoinflammatory gonarthritis. Ada Radiol 1994; 35: Tamai K, Yamato M, Yamaguchi T, Ohno W. Dynamic magnetic resonance imaging for the evaluation of synovitis in patients with rheumatoid arthritis. Arthritis Rheum 1994;8: stergaard M, Stoltenberg M, Henriksen O, Lorenzen I. Quantitative assessment of synovial inflammation by dynamic gadolinium-enhanced MRI. A study of the effect of intraarticular methylprednisolone on the rate of early synovial enhancement. Br J Rheumatol 1996; 35: Arnett FC, Edworthy SM, Bloch DA et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988;31: Polisson RP, Schoenberg OI, Fischman A et al. Use of magnetic resonance imaging and positron emission tomography in the assessment of synovial volume and glucose metabolism in patients with rheumatoid arthritis. Arthritis Rheum 1996;6: Winalski CS, Aliabadi P, Wright RJ, Shortkroff S, Sledge CB, Weissman BN. Enhancement of joint fluid with intravenously administered gadopentetatedimegluminc: Technique, rationale, and implications. Radiology 1993;187: Yamato M, Tamai K, Yamaguchi T, Ohno W. MRI of the knee in rheumatoid arthritis: Gd-DTPA perfusion dynamics. J Comput Assist Tomogr 1993;17: Drape J, Thelen P, Gay-Depassier P, Silbermann O, Benacerraf R. Intraarticular diffusion of Gd-DOTA after intravenous injection in the knee: MR imaging evaluation. Radiology 1993;188: Herve-Somma CMP, Sebag GH, Prieur A, Bonnerot V, Lallemand DP. Juvenile rheumatoid arthritis of the knee: MR evaluation with Gd-DOTA. Radiology 1992;182: Jevtic V, Watt I, Rozman B et al. The value of contrast enhanced magnetic resonance imaging in evaluation of drug therapy in rheumatoid arthritis A prospective study on hand joints in 65 patients. Ada Pharm 1993;43: Yanagawa A, Takano K, Nishioka K, Shimada J, Mizushima Y, Ashida H. Clinical staging and gadolinium-dtpa enhanced images of the wrist in rheumatoid arthritis. J Rheumatol 1993;20:781-4.