Rheumatoid arthritis (RA) is a systemic inflammatory disease that results in cartilage and bone destruction. RA is characterized by a typical pattern and distribution of synovial joint involvement. Disorganization of the joint leads to deformities and loss of function.
In the hands, the metacarpophalangeal (MCP), proximal interphalangeal (PIP), and thumb interphalangeal (IP) joints are most frequently involved. The distal interphalangeal (DIP) joints are involved only in the presence of a coexisting MCP or PIP disease. Tenosynovitis of the flexor tendons causes a reduction in finger flexion and grip strength. Nodular thickening in the tendon sheath may also produce a trigger finger.
In the wrist, the early stages of RA cause tenosynovitis of the extensor tendons, causing swelling over the distal wrist. The ulnar styloid may become tender, which indicates inflammatory synovitis. The distal end of the ulna tends to sublux dorsally, and the carpal bones sublux anteriorly to the distal radius and ulna. Bony erosions and ankylosis of the carpal bones are also seen and appear to be prominent features in Asian patients.
The primary effect of RA is in joint deformity and fusion, which occurs in the advanced stages.
Although occasional flares of joint pain occur throughout the course of the disease, these can usually be controlled with the use of anti-inflammatory medication, especially early in their course. When joint subluxations and deformity take place, performing basic daily tasks (eg, writing and holding utensils) can become a problem. Some patients resort to the use of custom-designed writing instruments or utensils to overcome this difficulty. Permanent disability occurs in approximately 10-20% of patients.
See the images below.
See Rheumatoid Arthritis: In and Out of the Joint, a Critical Images slideshow, to help identify the distinguishing features of RA as well as the signs of extra-articular manifestations of this disfiguring disease.
The primary role of imaging in RA of the hands and wrists is to exclude septic arthritis. Percutaneous ultrasonography-guided aspiration may be useful for localizing and obtaining samples of any collection that can be found. (See the image below.)
Radiography remains the first choice in imaging RA. Radiography is cheap, is easily reproducible, and allows easy serial comparison for assessment of disease progression. The main disadvantage is the absence of specific radiographic findings in early disease, since visualization of erosions may only be seen later. 
Magnetic resonance imaging (MRI) provides a more accurate assessment, as well as earlier detection of lesions. Magnetic resonance imaging (MRI) continues to develop as a treatment tool, with the main thrust being detection of early disease at a stage at which disease-modifying drugs can be used; however, the cost of the examination and the small size of the joints involved limit widespread use. With further experience and cheaper scans, MRI scanning for the treatment of RA may gain acceptance in the future. [2, 1, 3, 4, 5]
RA may occur as part of a mixed connective tissue disease in which clinical and radiologic appearances are not typical of one particular disease.
Usually, the earliest sign of RA is a periarticular soft-tissue swelling with a fusiform appearance. Normal fat planes may be obliterated, which occurs as a result of joint effusion, edema, and tenosynovitis. Juxta-articular osteopenia is another early sign, particularly during the acute inflammatory stage. Osteopenia subsequently becomes more generalized as the disease progresses.  See the images below.
Initially, joint spaces in the small joints of the hands show widening as a result of effusion; however, with cartilage destruction, joint spaces narrow. Erosions usually begin at the bare area of the joint not covered by cartilage, such as the intracapsular articular margins. Marginal erosions occur as a result of direct mechanical action of the hypertrophied synovium and granulation tissue. See the images below.
The forces of tendon imbalance and capsular contraction result in joint subluxation and malalignment. Characteristic appearances include the boutonnière (flexion at the PIP joint and extension at the DIP joint) and swan-neck (extension at the PIP joint and flexion at the DIP joint) deformities at the IP joints. Ulnar deviation is also commonly present at the MCP joints. MCP joints may also dislocate and have overriding ends (main en lorgnette or telescoping joints). Subluxations are progressive and may occur with or without the presence of bony erosions. Tendinous involvement of RA may result in rupture of the tendon, causing increased disability. If significant involvement of the scapholunate joint is present, laxity or rupture may cause rotatory subluxation of the scaphoid. See the images below.
Fusion or joint ankylosis is common in the later stages of RA. Fusion usually takes place in a deformed or malaligned position. This further reduces the functionality of the hand and affects independence in the activities of daily living. In the end stages, extensive erosions may combine to result in resorption and tapering of the ends of the bones. See the images below.
Complications of therapy may be seen, such as generalized osteoporosis from steroid usage. Large bony erosions and severe osteopenia predispose patients to relatively atraumatic fractures. Long-term steroid use also predisposes patients to avascular necrosis.
The hands are affected symmetrically, and the disease process is usually visible first at the second and third MCP joints and the third PIP joint. The disease continues in the remainder of the MCP and PIP joints. Varying stages may be seen in different joints at a single point in time. DIP joint involvement without proximal involvement is rare.
Erosions may be detected first either in the MCP and PIP joints or at the carpal bones. Erosions may also be seen at the intra-articular portion of the distal end of the radius or within the carpal bones. Carpal bone ankylosis is a common and fairly specific sign, particularly in the Asian population, in whom it tends to occur early in the disease process.
Degree of confidence
RA is indicated on radiographs by the overall pattern and features of joint involvement. Although the presence of bony erosions is typical of RA, erosions are not specific because of the large number of other erosive arthritides. An absence of osteophyte formation and subchondral sclerosis is also indicative of RA rather than osteoarthritis.
Of concern is the possibility of misdiagnosing another erosive arthropathy as RA or vice versa. Because the diagnosis of RA is a predominantly clinical one, the primary role of imaging is in follow-up observation.
In the late stage (but not in the early stage), the deformities and joint fusion of RA are usually distinguishable from other causes of erosive arthropathy. No normal variants simulate RA.
Currently, computed tomography (CT) has a minimal role in treating RA of the hands. Although CT is useful for demonstrating bony pathology, erosions in the small joints of the hands are evaluated best by using a combination of radiography and MRI. MRI also has the added advantage of being able to visualize bone marrow edema and synovial and soft-tissue involvement.
Magnetic Resonance Imaging
MRI provides images with good delineation of soft-tissue changes, cartilaginous defects, and osseous erosions associated with RA. In particular, the ability to detect synovial hypertrophy and pannus formation before the onset of bony erosions has become more valuable with the advent of disease-modifying antirheumatic drugs. The drugs, which retard the progression of RA, are most effective in the early stages of disease. [3, 4, 8, 9, 10] MRI has been shown to be more sensitive to early changes in RA, and in the appropriate clinical setting, it is more accurate than plain radiography in the diagnosis of RA.
Signal intensity of the inflamed synovial lining may vary markedly on T1- and T2-weighted images. An inflamed synovial lining usually demonstrates low signal intensity on T2-weighted images, and the images may be even darker if hemosiderin or a predominantly fibrous component is present within the lining. Tenosynovitis resulting from inflammation is seen as high–signal-intensity fluid on T2-weighted sequences; see the image below.
The intravenous (IV) administration of a gadolinium-based contrast agent allows better distinction of inflammatory soft-tissue changes and can help differentiate pannus from eburnation. In addition, rapid enhancement of proliferative inflammatory synovium has been reported, compared to minimal signal change in healthy soft tissue and bone. Fat suppression and fast dynamic scanning have also been used to increase the sensitivity of synovial imaging. [11, 12, 13]
Gadolinium-based contrast agents have been linked to the development of nephrogenic systemic fibrosis (NSF) or nephrogenic fibrosing dermopathy (NFD). These diseases have occurred in patients with moderate to end-stage renal disease after being given a gadolinium-based contrast agent to enhance MRI or magnetic resonance angiography (MRA) scans. NSF/NFD are debilitating and sometimes fatal diseases. Characteristics include red or dark patches on the skin; burning, itching, swelling, hardening, and tightening of the skin; yellow spots on the whites of the eyes; joint stiffness with trouble moving or straightening the arms, hands, legs, or feet; pain deep in the hip bones or ribs; and muscle weakness.
MRI is clearly superior to radiography in the early detection of bone erosions. MRI is more sensitive to bone marrow edema. Bright lesions on T2-weighted images often do not correspond to findings on radiographs or T1-weighted images. These lesions are believed to represent potentially reversible pre-erosive changes. [14, 15] See the images below.
Routine use of MRI in evaluating rheumatoid arthropathy in the hands and wrist has been limited by the relatively high cost, the difficulty in positioning disabled patients, and the inability to image several joints simultaneously.
Olech et al evaluated the sensitivity and specificity of MRI in detecting erosions, bone edema, and synovitis in the metacarpophalangeal and wrist joints for RA by comparing scans of bilateral hands and wrists of 40 healthy subjects with those of 40 RA patients using 0.2 T extremity-MRI. A total of 3,360 bones were evaluated, and patients with RA were found to have significantly more erosions, as well as higher scores for bone edema and synovitis. Although age had a significant effect on the number of erosions in both groups, age became insignificant in RA patients when disease duration was factored in. The number of erosions correlated with positive rheumatoid factor and higher C-reactive protein values. Bone marrow edema was the most specific MRI lesion for RA, with 65% sensitivity and 82.5% specificity. 
In a study by Lisbona et al, subclinical inflammation was identified by MRI in 96.4% of patients with RA who had achieved sustained clinical remission. Significantly higher scores of bone marrow edema were observed after sustained remission in patients with progression of erosions, as compared to patients with no erosion progression. According to the authors of the study, persistence of higher scores of BME may explain the progression of bone erosions in patients with persistent clinical remission. 
According to one study, using MRI with maximum intensity projection (MIP) images, together with palpation, makes detailed evaluation of synovitis of the hand in RA easy. The researchers concluded that MIP images may predict further joint damage because they allow semiquantitative estimation of the degree of thickening of the synovial membrane. 
Contrast-enhanced MRI with maximum intensity projection (MRI-MIP) has been found to be a useful imaging technique to evaluate synovitis in hands with RA. 
Imaging of early synovial disease by MRI makes early detection possible, but the findings are not specific in differentiating RA from other causes of synovitis or joint erosions.
Schwenzer et al investigated dynamic contrast-enhanced MRI in the differential diagnosis of psoriatic arthritis (PA) and RA in the hand and wrist by evaluating 45 patients (31 with RA; 14 with PA) with a 3-T whole-body MR unit. There was a statistically significant difference observed between the 2 groups regarding relative synovial enhancement after 15 minutes following contrast injection; however, there was no difference in relative enhancement at 35 seconds, 52 seconds, or 3 minutes. Significant correlations were found between inflammatory parameters and dynamic contrast-enhanced parameters in patients with RA but not in those with PA. 
High-resolution sonography with a high-frequency probe is used for evaluation of the small joints in RA; see the image below. Joint effusion is hypoechoic, while the hypertrophic synovium is more echogenic. Rheumatoid nodules are seen as fluid-filled round cavities with sharp borders. Bone erosions may be seen as irregularities in the hyperechoic cortex. [21, 22]
Complications of RA, such as tenosynovitis and tendon rupture, can also be visualized by using ultrasonography. This is most useful in the MCP and IP joints. The carpal bones and the carpometacarpal joints are not visualized as well because of their irregular configuration and deeper location. [23, 24, 25, 26, 27, 28]
Sonography has been applied to the assessment of RA with the goal of improving on the current standard of conventional radiography. Ultrasonography, especially when augmented by amplitude color Doppler (ACD) imaging, has also provided clinically useful information in the assessment of RA. Originally, ACD or power Doppler scanning was developed as an improvement to existing vascular sonographic imaging techniques to delineate stenoses by imaging blood flowing in vessel lumina by virtue of higher sensitivity to flow and lack of directional dependence. 
ACD imaging has been applied to RA with the goal of evaluating the manifestation of hyperemia in the inflammatory joint tissues at the symptomatic sites of disease. Synovial hyperemia is a fundamental pathophysiologic feature of RA. Most likely, synovial hyperemia is the principal factor underlying the hallmark x-ray observation of periarticular demineralization on radiographs of patients with RA, and it is believed to occur in proportion to joint-destructive disease activity. Hyperemia in the tendon sheath is common as well. See the images below.
Two reports in the literature have explored a possible role for Doppler ultrasonography in evaluating arthritis, especially RA, on the basis of synovial hyperemia. Newman et al  used a 3-level semiquantitative approach by trained radiologists to grade Doppler flow subjectively in patients with RA, psoriatic arthritis, and calcium pyrophosphate deposition disease arthritis. The investigators found that the degree of abnormal tissue hyperemia visualized correlates grossly with the severity of local disease activity as measured by synovial-fluid white blood cell (WBC) counts and patients' subjective estimates of symptoms. The report did not include joint examinations or other physical/functional measurements.
Hau et al  from the University Hospital of Würzburg, Germany, reported findings from ultrasound in the MCP and PIP joints of 34 patients with RA and 15 control subjects over a 6-month period. In the protocol, each patient was examined only once. Patients were stratified into 3 groups based on swelling and tenderness. Ultrasonography was performed by using a Siemens Doppler unit and a 5- to 13-MHz linear-array transducer operating at 12 MHz. The authors identified joint pannus on gray-scale images as hypoechoic soft tissue adjacent to the articular surfaces and described the pannus distribution as favoring the radial side of the joints. Velocity color Doppler (ie, conventional direction-sensitive Doppler) imaging was used to evaluate vascularization and blood flow in the rheumatoid pannus as a surrogate marker of inflammation. The quantitation method used was analogous to that used by one of the coauthors of this article in preliminary studies.
A simple image-based quantification was obtained in the study by Hau et al  by summing the color pixels within all joints in each patient classification after selection of regions of interest within each joint. The result was a unitless numerical scale that showed a greater than tenfold difference between patients with active RA and controls. From the preliminary investigations, both the Newman and Hau studies [30, 31] concluded that power Doppler ultrasonography can reflect disease activity in RA, and each group suggested further research in this area.
Since 1999, several research groups have gone further in their development of power Doppler ultrasonography and its application to RA. Teh et al  evaluated quantitative power Doppler ultrasonography for the assessment of therapeutic response in rheumatoid synovitis in 13 consecutive patients over 7 months, with correlation to serum inflammatory markers and clinical evaluation. Quantitiative power Doppler spectra were derived within select regions of interest from each patient’s scan, both before and after treatment; essentially, the method used was an automated pixel-counting technique similar to that of Hau. There was extremely good correlation between the image-based quantitation of synovial hyperemia and the clinical and laboratory assessment of disease activity in this study.
Thus, quantitative power Doppler ultrasonography was able to accurately depict the patients’ response to treatment. The authors of the study felt that the addition of quantitative power Doppler ultrasonography to the routine assessment of therapeutic response in RA yielded more accurate findings than existing clinical and laboratory methods alone.
Strunk et al  compared quantified power Doppler ultrasonography with laboratory measurement of serum vascular endothelial growth factor (VEGF level) as markers for disease activity, finding a very high correlation (p< 0.0001) for ultrasound and no relation with the immediate serum VEGF level in the same patient. That same year, Salaffi et al  evaluated quantified power Doppler ultrasonography augmented by the use of IV ultrasound contrasts, such as Levovist, using a quantitative time-intensity curve integration to arrive at the quantification in arbitrary units. The results showed a correlation of the Doppler signal to other measures of disease and treatment efficacy.
Fukae et al studied the relation between synovial vascularity, assessed by quantitative power Doppler ultrasonography (PDUS), and structural bone damage in a single finger joint by comparing 190 metacarpophalangeal (MCP) and 190 proximal interphalangeal (PIP) joints of 19 patients with active RA who received disease-modifying antirheumatic drugs (DMARDs). The authors found that quantitative PDUS was significantly correlated with enhancement rate of MRI in each single finger joint. Comparing quantitative synovial vascularity and radiographic change in single MCP or PIP joints, the level of vascularity at baseline showed significant positive correlation with radiographic progression at the 20th week. Quantitative PDUS was found to be more useful than the semiquantitative method in evaluating synovial vascularity in a single finger joint. 
Fiocco et al  found that both C-reactive protein and Doppler findings reliably measured response to treatment with etancercept, an anti–tumor necrosis factor-alpha medication that has shown considerable efficacy in rheumatoid arthritis.
Research in this area is ongoing, but according to published reports, quantified power Doppler ultrasonography is a reproducible and reliable surrogate indicator of disease activity, which can predict and evaluate treatment response with a greater degree of sensitivity and specificity than other available clinical methodologies. [37, 38]