Ultrasonography is an elegant tool for the detection of tenosynovitis, synovitis, and erosions very early in rheumatoid arthritis, and the presence of a power Doppler signal is one of the best predictors of joint damage. Although clinical scores remain the mainstay of disease activity assessment, ultrasonography has proved to be a remarkably robust tool for reliable assessment of changes in rheumatoid arthritis. There is no evidence to suggest that problems with operator dependence would be greater than with other imaging modalities or physical examination, if performed by trained providers.
- •
High-frequency ultrasonography allows detailed assessment of superficial structures including tendons, tendon sheaths, joint capsule, cartilage, and cortical surface of bone.
- •
Gray-scale ultrasonography can visualize proliferative synovial tissue, fluid collections in tendon sheaths or joints, and bony erosions.
- •
Doppler ultrasonography can visualize synovial hyperemia, the strongest predictor of future joint damage in rheumatoid arthritis.
- •
Change in synovial thickening and hyperemia in response to treatment can be documented with serial ultrasonography.
Ultrasonography can now be a point-of-care modality: The provider performs the examination in the office, with no referral needed. Findings can be addressed immediately, and necessary adjustments of treatment can be made at the same visit. If a fluid collection is detected and an aspiration is indicated, ultrasound guidance can help improve the accuracy of the aspiration and injection.
Rationale for using ultrasonography in the assessment of early rheumatoid arthritis
Without imaging of soft tissues, providers in rheumatology have to rely on surrogate markers of joint inflammation. It is assumed that tenderness and swelling over joints are due to synovitis. If serologic markers of inflammation are abnormal, this is taken as an additional indicator of joint inflammation. However, fullness and pain on examination may have causes other than synovitis, and elevated sedimentation rates or levels of C-reactive protein may not always be due to RA. Even fibromyalgia patients may complain about morning stiffness and swollen hands.
Assessment of Early Rheumatoid Arthritis: Ultrasonography Versus Clinical Scores
Clinical scores remain the mainstay of assessment of disease activity. These scores, alone or in combination with measurement of acute-phase reactants, give no actual information about presence or absence of features of RA such as synovitis, tenosynovitis, or erosive disease, but serve as surrogate markers to help guide treatment recommendations. With ultrasonography or magnetic resonance imaging (MRI), synovial inflammation can often be detected in patients assumed to be in remission based on clinical scores alone. There thus appears to be a disconnect between Disease Activity Score (DAS) remission and imaging remission.
Interobserver and intraobserver reliability of ultrasonography and clinical scores have been compared in several studies. Ultrasonography was found to be at least as good, or better, for repeatability (intraobserver) and reproducibility (interobserver) of findings than was assessment of disease activity with surrogate clinical scores. When ultrasonography assessment is added to clinical scores, such a composite index (US-DAS) may be a better predictor of future joint damage than a standard DAS28 alone.
Assessment of Early Rheumatoid Arthritis: Ultrasonography Versus Conventional Radiography
In a world where conventional radiography is the only readily available imaging modality, the features of RA that can be assessed with this modality assume great importance. Erosions, joint-space narrowing, and periarticular osteopenia are the classic features. However, even though conventional radiography is the historical gold standard of erosion assessment, it is insensitive when compared with ultrasonography.
When ultrasonography was compared with conventional radiography in a study of patients with RA, sonography detected 6.5-fold more erosions in early disease, in 7.5-fold the number of patients ( Fig. 2 ). Two independent operators performed scans of metacarpophalangeal (MCP) joints sequentially for this study, and reached good interobserver reliability (Cohen κ = 0.75).
Is there such a thing as “operator dependence” when obtaining conventional radiographs? As radiography projects 3-dimensional structures on 2-dimensional planes, breaks in the bony cortex need to be seen in profile to be characterized as erosions (lesions that are not seen in profile may be either cysts or erosions). To help detect erosions of the cortex in RA, extremities are assessed in different views, for example, anterior-posterior, lateral and oblique, or “ball-catcher’s” views of the hands. Detection of cortical breaks will vary with positioning of the extremity. The radiology technician, and the patient, will make an effort to standardize the position of the extremity as much as possible. Nevertheless, a few degrees in difference of rotation can render an erosion impossible to detect. Furthermore, for an assessment at a different time point, for example, to evaluate progression of erosions, the exact same position would need to be assumed, which is problematic in a free-hand approach ( Fig. 3 ). Once radiographs are obtained, will different readers come to the same conclusions? John T. Sharp found that “the variability in scoring radiographic abnormalities is considerable among this group of 11 expert readers” in a study with standard data sets of patients with RA.
Synovitis or tenosynovitis, both very early findings in RA, cannot be assessed radiographically.
Assessment of Early Rheumatoid Arthritis: Ultrasonography Versus MRI
For the image assessment of RA, ultrasonography is often compared with MR imaging. Both are cross-sectional modalities that evaluate similar features of RA. It is therefore of considerable interest to identify the modality that is more sensitive, specific, reproducible, and repeatable. Both modalities can assess bony erosions, synovial tissue proliferation, synovial fluid collections, tenosynovitis, enthesitis, and synovial hyperemia. MRI can identify bone marrow edema, which ultrasonography cannot. Ultrasonography can be a dynamic, real-time examination, which MRI is not (dynamic cardiac MRI would be the exception). Displaceability of synovial fluid and compressibility of synovial tissue can only be assessed sonographically. Synovial hyperemia can be seen in real time with ultrasonography. The greatest extent of hyperemia during systole can safely be appreciated and used for scoring of disease. It has not been studied well if MRI assessment of hyperemia is subject to variation during systole and diastole, and if an MR image taken during diastole could lead to inaccurate assessment of synovial hyperemia. It is frequently mentioned that operator dependence could be a discriminating factor between MRI and ultrasonography. The operator dependence of MRI assessment of rheumatic disease has not been systematically studied. For such a study, 2 or more operators (radiology technicians or physician providers) would need to separately perform MRI studies of the same patients to assess predefined features of RA, using 2 or more different MRI coils. The resulting images would need to be read by 2 or more radiologists. If the agreement were 100% at all times, MRI could be called operator independent—an unlikely scenario at best. Unsurprisingly, the diagnostic accuracy of MRI interpretations varies between readers and depends on the experience of the provider. In a study of 21,482 consecutive computed tomography (CT), MRI, and ultrasonography studies, MRI and CT interpretations had significantly higher rates of discrepancy among readers than ultrasonography.
MRI and ultrasonography: erosions
A recent systematic literature review found no statistical difference in the efficacy of MRI and ultrasonography in detecting erosions in RA. In early RA, ultrasonography tended to detect more erosions than MRI, whereas MRI tended to detect more erosions in late disease. Good reproducibility of ultrasonography findings for erosion detection was found. Ultrasonography assessment of erosions may therefore be regarded more cost effective than MRI for this indication.
MRI and ultrasonography: synovitis and tenosynovitis
For a reliable assessment of RA, an imaging modality must depict the anatomic reality as faithfully as possible. Ultrasonography findings of synovial hyperplasia and hyperemia correlate well with histopathological findings after surgery. In a recent study, findings of power Doppler ultrasonography were closely associated with all “pathologic compartments” of synovitis, including inflammatory cell infiltrates, synovial lining layer thickness, and vascularity. Ultrasonography findings more faithfully illustrated active synovitis than those of MRI.
Small fluid collections and tenosynovitis in patients with RA are more readily detected by ultrasonography when compared with MRI.
Ultrasonographic features of rheumatoid arthritis
Tenosynovitis
There is some evidence that tenosynovitis is a very early feature of RA. It may precede synovial proliferation within the joints. Tenosynovitis can, by definition, only occur at sites where tendon sheaths encase the tendons. Tendon sheaths around the wrists particularly are affected early on but also later in the disease process. Tendon sheaths of the extensor carpi ulnaris tendon (extensor compartment 6) and extensor digitorum together with extensor indicis tendon (extensor compartment 4) are typical tendon sheaths affected by tenosynovitis in RA.
Sonoanatomy of tendon and sheath
Tendons with their sheaths share anatomic characteristics with joints. The tendon fibers are encased by a layer of synovial tissue, the inner lining or visceral layer. The fibrous tendon sheath connects to the adjacent periosteum and is lined by the outer or parietal synovial layer. The 2 synovial layers connect through a duplication along the mesotendineum, or mesotenon, a vascularized suspensory ligament that connects the tendon with its environment. In RA, synovial tissue can proliferate from both layers, as well as from the lining of the mesotendineum, the “third layer” ( Fig. 4 ).
Tenosynovial effusions
Fluid collections, or tenosynovial effusions, will distend the layers. Synovial fluid is usually anechoic. It will provide contrast to better appreciate the anatomic structures. In RA, effusions of the tendon sheath are often associated with synovial proliferation. By contrast, effusions caused by mechanically induced tenosynovitis will have only little coexisting synovial thickening.
Synovial proliferation
The delicate synovial lining of tendon and sheath can usually not be seen sonographically in healthy individuals. Detection of synovial tissue is an abnormal finding. Synovial tissue will appear as hypoechoic, gray-appearing tissue interposed between the fibrous and bright, or hyperechoic-appearing tendon sheath and the bright, hyperechoic tendon fibers. If synovial fluid is present, it will usually appear as dark or anechoic, and will accentuate the anatomy of tendon, sheath, and synovial tissue. With transducer pressure, fluid will be displaceable: it will flow away from the pressure of the transducer. By contrast, synovial tissue is not displaceable and will stay in place. A minimal compressibility of synovial tissue may be seen, due to its often villous structure with sponge-like qualities.
Synovial hyperemia
No significant blood flow is usually detectable sonographically in tendon and sheath of healthy adults. If synovial proliferation is detected or suspected on gray-scale ultrasonography, this tissue must be examined for hyperemia using Doppler ultrasonography.
Paratenonitis
Another concept is the idea of paratenonitis. Inflamed tissue adjacent to tendons without sheath can occasionally be seen by ultrasonography in inflammatory arthritis, including RA. This feature has found its way into the US7 ultrasonography scoring system for RA. It is unclear what precise tissues become edematous and hyperemic adjacent to sheathless tendons, and paratenonitis has not been described elsewhere as a typical feature of RA. It is possible that synovial pannus breaks through the joint capsule and allows for synovitis to distribute along the tendon ( Fig. 5 ).
Synovitis
Based on clinical examination alone, it may be difficult to tell if palpable fullness or warmth over a joint, or tenderness on palpation, is due to subcutaneous edema, tenosynovitis, paratenonitis, a joint effusion, or synovial proliferation within the joint. Ultrasonography can help with a more precise assessment.
Joint effusion
Ultrasonography appears to be more sensitive than clinical examination or MRI for the detection of small, supraphysiologic fluid collections in early RA. Synovial fluid has an anechoic, or “black” sonographic appearance. It will distend the more hyperechoic fibrous joint capsule. Early on, synovial fluid will collect in the proximal recesses of the joint capsules of small joints of hands and feet ( Fig. 6 ). Later in the disease process, excess synovial fluid can displace the fibroadipose pad that reinforces the dorsal joint capsule in these joints, and eventually lead to a convex elevation of the joint capsule over proximal and distal bony endings. The fibrous volar and plantar plates that prevent overextension of finger and toe joints may be displaced by synovial fluid. In contrast to tissue, synovial fluid moves freely within the joint. Ultrasonography is a dynamic, real-time imaging modality. Transducer pressure on the joints can displace fluid away from the transducer. By contrast, synovial tissue will be just slightly compressible (sponge-like) as in tendon sheaths, but will not be displaced by transducer pressure.
Synovial hypertrophy
Proliferation of synovial-lining tissue will be one of the earliest features of RA that can be detected sonographically. As physiologic synovial lining tissue is only 1 to 3 cell layers strong, any sonographically detectable synovial thickening raises the suspicion of inflammatory arthritis. Sonographically, synovial tissue will appear hypoechoic or gray, in contrast to adjacent hyperechoic, fibrous capsule tissue; anechoic, or black-appearing synovial fluid; anechoic hyaline cartilage; and hyperechoic bony contour.
Synovial hyperemia
Once synovial tissue is detected using gray-scale, or B-mode, ultrasonography, color Doppler or power Doppler can help assess the degree of hyperemia of this tissue ( Fig. 7 ).
Doppler signals will represent blood flow rather than artifact if they are seen over an area that was identified as abnormal in the preceding gray-scale assessment, have a pulsatile quality synchronous to the pulse, and are seen persistently in the same area during a real-time examination.
Bony Erosions
Bony erosions can be detected sonographically relatively early in the disease process ( Fig. 8 ). Ultrasonography of erosions performs better in joints that are readily accessible to sonographic evaluation, including small joints of hands and feet. It is less strong in deeper-seated bony structures and in bones that will not allow circumferential assessment, such as the carpal bones in wrists or the tarsal bones in feet. Bony erosions are defined as breaks in the bony cortex, seen in 2 perpendicular planes. It is helpful to gain experience with the sonographic appearance of the anatomic neck of the distal metacarpal and metatarsal bones, as it is located near typical sites of erosion in RA.
Related posts:
Genetics of Rheumatoid Arthritis: An Impressionist Perspective
Evolution of Classification Criteria for Rheumatoid Arthritis: How Do the 2010 Criteria Perform?
Early Spondyloarthritis
Magnetic Resonance Imaging Applications in Early Rheumatoid Arthritis Diagnosis and Management
Early Juvenile Idiopathic Arthritis
Early Psoriatic Arthritis
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
