Musculoskeletal ultrasound has emerged as a key tool for the diagnosis, prognosis, and management of patients with RA (rheumatoid arthritis) and other rheumatic diseases. The most important sonographic findings in RA include erosions, effusions, synovitis, and tenosynovitis. Investigators have suggested various “optimal” numbers of joints to scan in RA to assess disease activity, gauge treatment response, provide prognostic information, and guide management decisions. The complexity of pediatric sonoanatomy has delayed its validation in juvenile idiopathic arthritis, yet ultrasound reliably measures the extent of synovitis/tenosynovitis and guides precise injections.
Key points
- •
Musculoskeletal ultrasound has emerged as a key tool for the diagnosis, prognosis, and management of patients with RA (rheumatoid arthritis) and other rheumatic diseases.
- •
The most important sonographic findings in RA include erosions, effusions, synovitis, and tenosynovitis.
- •
Investigators have suggested various “optimal” numbers of joints to scan in RA to assess disease activity, gauge treatment response, provide prognostic information, and guide management decisions.
- •
The complexity of pediatric sonoanatomy has delayed its validation in juvenile idiopathic arthritis, yet ultrasound reliably measures the extent of synovitis/tenosynovitis and guides precise injections.
Introduction
The field of rheumatology has made great strides over the past 2 decades in understanding the pathogenesis of rheumatoid arthritis (RA), as well as developing biologic treatments that have fundamentally altered the natural history of the disease. These emerging therapeutic options in part spurred the 2010 update of the RA classification (and diagnostic) criteria, which stresses the importance of early identification and aggressive treatment of RA, not only for alleviating symptoms but also for altering the disease process and preventing damage. Yet these criteria, as well as measures of disease activity (HAQ-DI, DAS-28) and serum markers of inflammation (erythrocyte sedimentation rate [ESR] and C-reactive protein [CRP]), often are not sufficiently reliable in identifying patients with RA or assessing their disease activity. Moreover, these tests are oftentimes difficult for patients to understand in meaningful ways.
Perhaps addressing these shortcomings, rheumatologists overseas (and more recently in the United States) have increasingly looked to musculoskeletal ultrasound (MSUS) to aid in the diagnosis and management of RA and other types of arthritis. Although other imaging modalities provide helpful information about RA, MSUS holds many advantages over plain radiography and magnetic resonance imaging (MRI), and can be performed during the clinical visit by ultrasound-trained rheumatologists ( Box 1 ). Not only can MSUS aid in the diagnosis of RA in difficult cases, but many studies have demonstrated that it can provide prognostic insight and have implications for treatment decisions. For instance, a 2008 study by Brown and colleagues (presented later in more detail) identified patients in clinical “remission” with subclinical inflammation by MSUS that predicted ongoing structural damage. In their study, the conventional measures of disease activity often lacked sensitivity to identify the need for more aggressive treatment, whereas the presence of synovitis, as defined by MSUS, was highly associated with progressive radiographic changes.
No complications
No ionizing radiation
No claustrophobia, anxiety, or need for sedation or anesthesia
No contraindications (metal implants or pacemakers)
Ability to generate dynamic images
Ability to assess multiple joints in one session
Relatively low cost
As the excitement about MSUS has spread to other disorders within rheumatology, assisting in the diagnosis and treatment of crystal disease, spondyloarthropathies, osteoarthritis, and others, the most robust and organized investigations appear in the RA and juvenile idiopathic arthritis (JIA) literature. The results of these studies have, in many ways, ushered in a Renaissance in rheumatology, questioning previously “accepted” methods of examination and changing the way rheumatologists think about arthritic disease. We will show that in the past 5 years, sonographers have proposed various MSUS scoring systems for use in RA research studies and in clinical practice, yet a gold standard consensus for using this imaging modality in RA treatment algorithms remains elusive.
Introduction
The field of rheumatology has made great strides over the past 2 decades in understanding the pathogenesis of rheumatoid arthritis (RA), as well as developing biologic treatments that have fundamentally altered the natural history of the disease. These emerging therapeutic options in part spurred the 2010 update of the RA classification (and diagnostic) criteria, which stresses the importance of early identification and aggressive treatment of RA, not only for alleviating symptoms but also for altering the disease process and preventing damage. Yet these criteria, as well as measures of disease activity (HAQ-DI, DAS-28) and serum markers of inflammation (erythrocyte sedimentation rate [ESR] and C-reactive protein [CRP]), often are not sufficiently reliable in identifying patients with RA or assessing their disease activity. Moreover, these tests are oftentimes difficult for patients to understand in meaningful ways.
Perhaps addressing these shortcomings, rheumatologists overseas (and more recently in the United States) have increasingly looked to musculoskeletal ultrasound (MSUS) to aid in the diagnosis and management of RA and other types of arthritis. Although other imaging modalities provide helpful information about RA, MSUS holds many advantages over plain radiography and magnetic resonance imaging (MRI), and can be performed during the clinical visit by ultrasound-trained rheumatologists ( Box 1 ). Not only can MSUS aid in the diagnosis of RA in difficult cases, but many studies have demonstrated that it can provide prognostic insight and have implications for treatment decisions. For instance, a 2008 study by Brown and colleagues (presented later in more detail) identified patients in clinical “remission” with subclinical inflammation by MSUS that predicted ongoing structural damage. In their study, the conventional measures of disease activity often lacked sensitivity to identify the need for more aggressive treatment, whereas the presence of synovitis, as defined by MSUS, was highly associated with progressive radiographic changes.
No complications
No ionizing radiation
No claustrophobia, anxiety, or need for sedation or anesthesia
No contraindications (metal implants or pacemakers)
Ability to generate dynamic images
Ability to assess multiple joints in one session
Relatively low cost
As the excitement about MSUS has spread to other disorders within rheumatology, assisting in the diagnosis and treatment of crystal disease, spondyloarthropathies, osteoarthritis, and others, the most robust and organized investigations appear in the RA and juvenile idiopathic arthritis (JIA) literature. The results of these studies have, in many ways, ushered in a Renaissance in rheumatology, questioning previously “accepted” methods of examination and changing the way rheumatologists think about arthritic disease. We will show that in the past 5 years, sonographers have proposed various MSUS scoring systems for use in RA research studies and in clinical practice, yet a gold standard consensus for using this imaging modality in RA treatment algorithms remains elusive.
Historical perspective of MSUS in rheumatoid arthritis
One of the earliest reported cases of a patient with RA evaluated by using MSUS was a 1972 report of a patient with a large Baker’s cyst, which was clinically diagnosed as thrombophlebitis. In 1978, Cooperberg and colleagues speculated on the usefulness of MSUS in RA by its ability to better characterize suprapatellar effusion sizes, popliteal cysts, and synovial thickening. It was not until the 1990s that the body of literature evaluating RA with MSUS began to expand. A 1993 study by Grassi and colleagues identified significant sonographic findings in RA (described later in more detail), including erosions, synovitis, and joint effusions. In 1995, Lund and colleagues also described the articular abnormalities of the metacarpophalangeal (MCP) as well as carpal joints in 29 patients with RA and were able to correlate these findings with disease severity based on clinical assessment.
Although studies demonstrated the ability of MSUS to identify both articular and periarticular disease in RA, subsequent efforts were aimed at comparing MSUS with clinical examination, conventional radiography, and MRI ( Table 1 ). A 1999 prospective study of 60 patients with inflammatory arthritis (36 with RA) looked at MSUS versus plain radiography, bone scintigraphy, and MRI, concluding that both MSUS and MRI were better able to identify both tenosynovitis and erosive changes in patients without apparent radiographic abnormalities. In 2005, Naredo and colleagues studied 94 patients with RA to demonstrate that MSUS was superior to clinical examination in evaluating disease activity in terms of effusions and synovitis, and that the intraobserver/interobserver reliability was similarly more robust with MSUS than with clinical examination. Since then, several other studies have shown similar findings comparing MSUS to clinical examination, and have also shown MSUS to reveal erosions earlier in the disease process than does plain radiography. Numerous studies have also found that MSUS can usually detect erosions and synovitis, as well as (and sometimes better than) the more expensive and less accessible MRI tests ( Table 2 ). One report by Swen and colleagues evaluating finger tenosynovitis in 21 patients with RA found that the sensitivity and specificity for the determination of partial tendon tears was slightly higher using MSUS compared with MRI. Although the precise role for MSUS in RA management was not yet defined, these earlier studies at least started the discussion in defining its proper place among other useful imaging modalities in RA.
| Erosions | Synovitis | Tenosynovitis | Peri-Articular Osteopenia | |
|---|---|---|---|---|
| Radiograph | Yes | No | No | Yes |
| Magnetic resonance imaging | Yes | Yes | Yes | Yes |
| Ultrasound | Yes | Yes | Yes | No |
| Synovitis | Erosions | |||||
|---|---|---|---|---|---|---|
| Sensitivity (%) | Specificity (%) | Sensitivity (%) | Specificity (%) | |||
| Szkudlarek et al, 2006 | PE | 40 | 85 | XR | 42 | 99 |
| US | 70 | 78 | US | 59 | 98 | |
| Wakefield et al, 2008 | PE | 55–83 | 23–46 | N/A | ||
| US | 64–89 | 60–80 | ||||
| Rahmani et al, 2010 | N/A | XR | 13 | 100 | ||
| US | 63 | 98 | ||||
| Alarcon et al, 2002 | N/A | MRI | 100 | 65 | ||
| US | 100 | 45 | ||||
Key sonographically defined pathology in RA
The identification of certain sonographic features of RA has become particularly useful in both clinical practice and research studies. The salient findings include erosions, effusions, synovitis, and tenosynovitis. Before 2004, there was a scarcity of both reliability and validity data using MSUS, as well as little consensus around definitions of pathologic findings of inflammatory arthritis. In response to this, a group of international sonographers created the OMERACT Ultrasound Task Force to address these deficiencies. Since then, the task force has succeeded in creating standardized definitions for various pathologic findings, as well as improving reliability in the detection of synovitis.
Erosions
Although erosive changes are not a pathognomonic feature of RA, the ability to detect these lesions in the right setting can help to establish a diagnosis, as well as predict more aggressive disease and worse clinical outcomes, justifying more aggressive therapy. Identification of erosions also facilitates enrollment into studies and contributes to outcome measures. Plain radiography at regular intervals was previously used to detect erosions in RA, but this uniplanar technique misses erosive changes early in disease, and damage in patients with superimposed secondary degenerative changes. By contrast, MSUS has been shown to detect erosive changes months or years earlier in the disease process and can visualize articular changes in multiple planes of view.
The OMERACT group defines the sonographic bone erosion in RA as “an intra-articular discontinuity of the bone surface that is visible in 2 perpendicular planes.” Although erosions can be visualized in many anatomic locations, the most likely areas for detecting hand erosion formation using MSUS include the ulnar styloid process, radial aspect of the second MCP joint, and ulnar aspect of the fifth MCP joint. Because of limitations in probe positioning, MSUS is less useful in evaluation of the carpal bones.
Several studies have shown the superiority of MSUS in the evaluation of erosions compared with conventional radiography. A 2000 article by Wakefield and colleagues, evaluating the MCP joints of 100 patients with RA, concluded that MSUS was able to detect 127 erosions in 56% of patients, compared with radiographic detection of 32 erosions in only 17% of patients ( P <.0001). Of note, a subgroup analysis of patients with early RA (disease duration less than 12 months) revealed that 6.5-fold more erosions were seen using MSUS compared with conventional radiography. Another study by Lopez-Ben and colleagues similarly concluded that MSUS surpassed radiography in the ability to detect more erosions in the hands and feet among patients with RA. Although some may defer to MRI for detecting erosions, many studies have shown that MSUS can detect erosions at a similar rate.
Importantly, high interobserver reliability in the detection of bone erosions using MSUS was observed between an experienced radiologist and a rheumatologist. Similar results were obtained more recently following a study in which a rheumatologist without prior experience in sonography underwent an intensive 4-week training course in MSUS and attained high interobserver rates for detecting bone erosions, as compared with a rheumatologist experienced in MSUS.
Effusions
Although pathologic collections of synovial fluid are seen by MSUS in many diseases, including knee osteoarthritis, they are common in almost any active RA joint, from the small hand and foot interphalangeal joints to the medium-sized wrists and larger shoulders and hips ( Fig. 1 ). OMERACT defines sonographic effusions as abnormal hypoechoic (gray) or anechoic (black) intra-articular material that is both displaceable and compressible, but lacks Doppler signal. MSUS can also quantify the effusion size and assist with both diagnostic and therapeutic arthrocentesis of difficult joints.
Synovitis/Tenosynovitis
OMERACT defines synovial hypertrophy as abnormal hypoechoic intra-articular tissue that is nondisplaceable and poorly compressible, which may exhibit Doppler signal. Proliferative synovitis, also known as pannus, is the hallmark pathologic lesion in RA and represents the primary site of inflammation. Gray-scale images alone for the evaluation of synovitis ( see Fig. 1 ) cannot always discern between acutely and chronically inflamed synovial membranes, as the latter can maintain its thickness over time and make it more difficult to determine therapeutic responses to treatment.
The added information from using power Doppler in addition to the gray-scale imaging information allows the clinician to determine the amount of vascularity within synovial tissue ( Fig. 2 ), which can be assessed either semiquantitatively (on a 0–3 scale) or quantitatively (summating pixel numbers in a region of interest). This technique helps assess synovitis in the gamut of joint sizes from the small joints of the digits to the shoulder and hip. In fact, assessment of vascularity using Doppler has been shown to correlate with the histopathologic features of synovitis. A study by Walther and colleagues that evaluated the knee joints of 23 patients with arthritis undergoing arthroplasty showed high correlation ( P <.01) between Doppler scoring of vascularity performed before surgery and histologic analysis of synovial tissue (by qualitative and quantitative vessel density in the sample) obtained during the procedure. Another study among patients with severe hip arthritis planning to undergo arthroplasty showed similar results, suggesting that Doppler reliably measures vascularity of synovial tissue.
Ultrasonography is also useful for the detection of tenosynovitis, defined sonographically by OMERACT as either anechoic or hypoechoic tissue with or without fluid in the tendon sheath, seen in 2 perpendicular planes and which may demonstrate Doppler signal. Common locations for tendon involvement in RA are in the hand and wrist, including the flexor digitorum, extensor digitorum ( Fig. 3 ), extensor carpi ulnaris, and extensor carpi radialis tendons. Recent studies suggest that evaluation of tenosynovitis using MSUS is reliable, with acceptable intraobserver and interobserver concordance.
Key pathology in other inflammatory arthritides
Some of the sonographic features seen in RA overlap with those found in other rheumatic diseases, thus it is important to note that they are not pathognomonic for RA. At the same time, some of the other diagnoses have other unique sonographic findings ( Table 3 ).
| Rheumatoid arthritis | Synovitis, effusions, erosions, tenosynovitis |
| Psoriatic arthritis | Synovitis, effusions, erosions, tenosynovitis, enthesitis |
| Gout | Synovitis, effusions, erosions, tophi, double contour (uric acid on articular cartilage) |
| Calcium pyrophosphate disease | Synovitis, effusions, calcium pyrophosphate deposition within articular cartilage, in fibrous cartilage and tendons or as freely mobile nodular deposits in bursae |
| Osteoarthritis | Synovitis, effusions, osteophytes, articular cartilage thinning |
Spondyloarthritides
Although psoriatic arthritis (PsA) and the spondyloarthropathies share some of the RA features detailed previously, both clinically and by ultrasound, evidence of entheseal changes represents one of the major differences, a reflection of its role as an initiating site of injury and subsequent inflammation. Such entheseal inflammation is often subclinical, although estimates of its asymptomatic prevalence in the PsA population vary widely. The enthesis represents 1 of the 5 main anatomic structures that can be abnormal sonographically in PsA, with others including the joint, tendon, skin, and nail. One group of investigators who outlined these 5 targets also created a preliminary Doppler score for PsA assessment that involves all 5 of them. Studies have consistently found MSUS to be superior to physical examination and radiograph in the detection of inflammatory changes. The utility of MSUS in management of PsA was demonstrated in a recent retrospective study that suggested that serial Doppler evaluations showed improvements in synovial proliferation and joint effusions (but not bone erosions) during the course of adalimumab therapy. The next step, however, remains the development of an accepted standardized scoring system for PsA assessment.
Crystalline Disorders
Although there have been numerous articles outlining the sonographic findings in crystal diseases, they too lack any universal scoring systems. Thiele and Schlesinger identified a “double-contour sign” of urate deposits on the hyaline cartilage in 92% of 23 patients with gout and in none of the control patients diagnosed with other arthritides. Subsequent studies confirmed greater than 98% specificity of the double-contour sign in gout, whereas Thiele and Schlesinger later found the double contour to disappear in all 5 patients who achieved a serum urate level less than 6 mg/dL for more than 6 months using urate-lowering drugs. Other investigators have also characterized sonographic findings of tophi and erosions demonstrating better sensitivity versus plain radiography. A recent study validated the inter-reader reliability of diagnosing tophi in the first metatarsophalangeal (MTP) joint and a double-contour sign in the femoral articular cartilage in patients with gout and asymptomatic hyperuricemia in a cohort of 75 patients.
MSUS can also detect calcium pyrophosphate disease in a number of different patterns and anatomic locations, including (1) hyperechoic bands within hyaline articular cartilage; (2) punctate hyperechoic spots, more common in fibrous cartilage and tendons; and (3) homogeneous, hyperechoic nodular deposits floating freely within bursae. Although its sensitivity varies widely in the literature, this MSUS finding in chondrocalcinosis appears to be highly specific.
Osteoarthritis
Although osteoarthritis (OA) has long been heralded as a noninflammatory, degenerative disease, MSUS findings are consistent with other clinical and imaging evidence that OA can in fact present with effusions and synovitis (by gray scale and Doppler). Beyond identifying osteophytes and articular cartilage wear, MSUS may serve a prognostic role in these patients. A recent prospective multicenter study of 531 European patients with painful knee OA identified predictive factors for knee replacement, one of which was an MSUS effusion measuring greater than 4 mm (while a clinically apparent effusion was not predictive).
Ultrasound scoring systems in RA
Although MSUS, as shown previously, is able to identify RA disease activity and damage at the “joint” level, many groups have devised methods of scoring at the “patient” level ( Table 4 ). Although no single scoring system has demonstrated superiority, the various systems aid in measuring clinical trial outcomes, help the clinician evaluate patient improvement (or worsening), and often provide important prognostic information, even in the case of subclinical inflammation. Sonographic examination of all joints of the hands and feet typically involved in RA can be too time-consuming and cumbersome, and thus impractical in the clinical setting and often for research, whereas systems scoring fewer joints or views may compromise overall sensitivity too much to be useful. The proposed number of joints to scan has ranged from 78 down to 6, although Grassi and colleagues recently proposed following only “the MCP with the most florid synovitis on initial US screening,” and aptly named this one-joint method as the SAS1 (Sonographic Activity Score) system.
| No. of Joints | Specific Joints Scanned | What Is Scored a | Time Needed | |
|---|---|---|---|---|
| Scirè et al, 2009 | 44 | Bilateral shoulders, elbows, wrists, MCPs, PIPs, sternoclaviculars, acromioclaviculars, knees, ankles, MTPs | Synovitis | About 60 min |
| Naredo et al, 2005 | 28 | Bilateral MCPs, PIPs, wrists, elbows, shoulders, knees | Effusion, synovitis | Maximum 30 min |
| Perricone et al, 2012 | 12 | Bilateral elbows, wrists, MCP2, MCP3, knees, ankles | Effusions, synovitis | Mean 22 min |
| Backhaus et al, 2009 | 7 | Wrist, MCP2, MCP3, PIP2, PIP3, MTP2, MTP5, all on the patient’s dominant side | Erosions, synovitis, tenosynovitis | 10–20 min |
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
