Diagnosis

Using the ACR classification criteria for diagnosing RA as a guide, ultrasound allows the examiner to detect joint inflammation, the symmetric characteristics of RA, and the range of joint involvement. One of the pioneer studies by Backhaus and colleagues performed on a group of 60 patients with different arthritides concluded that ultrasound could detect statistically significantly more inflammatory changes in the assessed finger joints ( Fig. 13-1 ) than clinical examination. In a study of finger (i.e., metacarpophalangeal [MCP] and proximal interphalangeal [PIP]) joints of 40 RA patients with early and established disease, Szkudlarek and coworkers used contrast-enhanced MRI as a reference method and showed that the sensitivity of detection for signs of inflammation with gray-scale ultrasound compared with clinical examination increased from 0.40 for clinical assessment to 0.70 for ultrasound. The difference was even more striking for metatarsophalangeal (MTP) joints. In a group of 40 RA patients with early and established disease, the sensitivity of assessment for signs of inflammation with clinical examination or gray-scale ultrasound increased from 0.43 to 0.87. In both studies, sensitivities increased without loss of specificity. In an article by Rees and associates, finger joints of 40 RA patients were assessed clinically with gray-scale ultrasound and with power Doppler ultrasound before and after intravenous contrast administration. Finding synovial vascularity in the joints that were clinically inactive, the authors questioned the traditional way of assessing disease activity. Moreover, Ellegaard and colleagues, presenting the results of assessment of wrist joints in 109 RA patients, suggested that examination of a single affected joint could be used as a measure of disease activity.

Synovitis in the second metacarpophalangeal joint is identified on the dorsal longitudinal view. A, Synovial thickening on gray-scale ultrasonography. B, Doppler signal in the thickened synovium.

Bone erosions ( Fig. 13-2 ) can be detected in areas accessible to ultrasound. In a heterogeneous group of 60 patients with different arthritides, Backhaus and colleagues concluded that ultrasound could visualize more erosions than radiography. With MRI as a reference method, Wakefield and coworkers showed that ultrasound detected more erosions than radiography, especially in cases of early disease. In this study, ultrasound detected erosive disease in 56 of 100 RA patients, compared with 17 with erosions detected on radiography. In a subgroup of 40 patients with early RA, 15 had erosions identified on ultrasound, compared with 2 who had disease detected on radiography. Other studies suggest that ultrasound can detect more erosions than radiography and MRI. In a study by Alarcón and associates of 10 RA patients with erosive disease in selected finger and toe joints, 8 had disease detected by ultrasound, 7 with MRI, and 3 with radiography. In studies of finger and toe joints performed on 40 RA patients by Szkudlarek and colleagues, the sensitivity of detection of erosions improved compared with radiography, with MRI used as a reference method, from 0.42 to 0.59 for the finger joints and 0.32 to 0.79 for the toe joints; sensitivity improved without a loss of specificity. When using CT as a reference for ultrasound assessment of the MCP joints of 17 RA patients, Døhn and coworkers showed high specificity and moderate sensitivity for detecting erosions, indicating that ultrasound detected more erosions than CT in the assessed joints.

A small bone erosion of the third metacarpophalangeal head of a patient with early rheumatoid arthritis is seen on the longitudinal ( A ) and transverse ( B ) dorsal views.

The erosion criterion also was studied for the shoulder and wrist joints. Hermann and associates assessed the humeroscapular joint in 43 RA patients with radiography, ultrasound, and MRI and concluded that ultrasound and MRI supplement the radiographic examination by visualizing more erosions in the joint. Alasaarela and colleagues assessed the shoulder joints of 30 RA patients with ultrasound and MRI. They suggested that the area could be visualized more effectively by ultrasound than MRI and that more erosions were detected with ultrasound.

The wrist joint is not as accessible as the shoulder for ultrasound examination. MRI and radiography detected more erosions than ultrasound at baseline and at 6 months after the study began. The study by Hoving and coworkers included 46 patients with newly diagnosed RA.

Characterization of subcutaneous swellings as rheumatoid nodules may also be important for the diagnosis, although it is most likely only of historic value. There are published studies on the differences between RA and spondyloarthropathies, but none has examined the value of ultrasound in the follow-up assessment of unspecified arthritis for development of RA.

With the technical progress of gray-scale ultrasound of soft tissues, the assessment of the normal synovium can present a challenge. In a study by Schmidt and associates of the joints of 100 healthy participants, the normal values for some scanning positions were established. Extensive work by Ellegaard and colleagues on the finger joints of 24 healthy participants showed that multiplication of scanning positions may lead to pathologic assessment of practically all assessed joints. The quality of equipment with Doppler assessment is a crucial factor. With high-end units, it is possible to detect flow in the normal synovium, depending on the size of the joint, whereas with medium- and low-end units, it is possible to detect only some pathologic flow. Cutoff points are not established, but the parameters of the machine must be considered when assessing inflammatory activity in RA patients.

No clear-cut studies on the value of ultrasound for diagnosing RA have been published. However, the increasing sensitivity of ultrasound in relation to clinical methods of assessment indicates that a set of ultrasound criteria for diagnosing RA will be established. Until then, physicians can conclude that use of ultrasound can complement the ACR classification criteria when establishing a diagnosis of RA.

Late, Nondiagnostic Signs of Disease

Synovial thickening and joint effusion ( Fig. 13-3 ) can be an expression of acute or chronic inflammation. Sometimes, differentiation between the two findings may be difficult, but Doppler ultrasound can detect small differences. However, the assessment of inflammatory activity depends on the sensitivity for flow of the ultrasound unit. Most of the joints affected by RA can be visualized with ultrasound, and information on localization and the extent of joint involvement can be used for assessment of disease extent and to guide intra-articular injections.

The dorsal longitudinal view reveals an effusion in the first metatarsophalangeal joint of a patient with rheumatoid arthritis. This is a frequent finding in patients with osteoarthritis and in healthy persons.

Hip

The hip joint is a difficult area for clinical examination. Involvement of the joint ( Fig. 13-4 ) can be displayed by an experienced ultrasonographer using a low-frequency transducer, and material for microscopic and bacteriologic analysis can be delivered with an ultrasound-guided puncture. Diagnosis of hip and groin conditions is facilitated by this noninvasive method, because ultrasound can visualize parts of the hip joint, muscles, and tendons in the region.

The longitudinal oblique view shows an intra-articular effusion in the anterior synovial recess of the hip joint of a patient with rheumatoid arthritis.

Small Joints

The acromioclavicular, sternoclavicular, elbow ( Fig. 13-5 ), wrist ( Figs. 13-6 and 13-7 ), finger, ankle, midfoot, toe, and temporomandibular joints are all readily available for ultrasound assessment and ultrasound-guided intra-articular treatment.

Ultrasound shows elbow synovitis in the posterior midsagittal view.

Synovitis of the radiocarpal and midcarpal joints of the wrist can be seen in the dorsal longitudinal view.

Synovitis over the styloid process of the ulna is revealed in the dorsal longitudinal view.

Tendons, Ligaments, and Nerves

Visualization of joints, tendons, and ligaments is a major advantage of musculoskeletal ultrasound over clinical examination. Direct insight into those structures can influence decision making, guide therapy, and broaden the knowledge on the nature of RA.

Assessment of tendons, ligaments, and nerves affected by RA is possible when their localization allows examination. Tenosynovitis ( Figs. 13-8 and 13-9 ), tendon ruptures, nerve entrapment syndromes, and entheseal inflammation caused by RA can be diagnosed and, when possible or indicated, can be treated with the help of ultrasound.

Tenosynovitis of the flexor carpi ulnaris tendon at the wrist in the palmar longitudinal ( A ) and transverse ( B ) views.

Ultrasound shows tenosynovitis of the third finger’s flexor tendon at the proximal interphalangeal joint in the palmar longitudinal view.

Disease Course Follow-up

Many research groups have emphasized the use of ultrasound during follow-up visits with RA patients. However, the multitude of assessment systems used makes it difficult to reproduce the results and apply them in daily practice.

Many published studies describe diminishing gray-scale synovial thickening in joints after treatment. The principle of assessment is well illustrated in a study by Ribbens and associates, in which the wrist, MCP, and PIP joints of 11 RA patients were assessed before and after 6 weeks of treatment with infliximab. Measurements of synovial thickening in the joints showed significant decrease after treatment and correlated with the clinical disease activity score. In the small joints of the hands and feet of 20 mostly RA patients, Filippucci and colleagues showed significant decreases in joint cavity widening after intra-articular steroid administration. Terslev and coworkers, in a study of 51 RA patients treated with intra-articular steroids, predominantly in the wrist joints but also in the small peripheral joints, showed a decrease of 31% of the pretreatment area of the synovial membrane displayed on gray-scale ultrasound after administration of the steroids. A 7-year follow-up performed by Scheel and associates on 16 RA patients showed a decrease in gray-scale synovitis (defined as joint effusion or synovial hypertrophy) during the study period.

Many research groups have performed studies of Doppler activity before and after treatment. Newman and colleagues described a decrease in synovial perfusion assessed with power Doppler ultrasound in the knees of eight RA patients after intra-articular administration of steroids. Stone and coworkers graded power Doppler signal on a four-grade scale in MCP joints before and after treatment with intravenous or oral steroids, showing a significant decrease in the intensity of the Doppler signal. Similar studies were done in the small joints of the hands and feet after administration of intra-articular steroids, with significant decreases in blood flow assessed with power Doppler ultrasound.

Administration of tumor necrosis factor alpha (TNF-α)–blocking agents results in significant decreases in vascularity as assessed with Doppler ultrasound. Hau and associates published a study on MCP joints of five RA patients treated with etanercept and assessed with color Doppler ultrasound. The number of color signals per region of interest was estimated and displayed a significant decrease after 1 month’s treatment. Terslev and colleagues also studied the effect of treatment with etanercept on the number of color pixels in the region of interest—the color fraction (CF)—in 11 RA patients and found a significant decrease after 2 weeks from the beginning of the study, which was not maintained after 1 year. Moreover, an increase in peripheral resistance index (RI) occurred, and it was seen after 2 weeks and 1 year from the study’s start. The CF and RI values for a group of 29 RA patients with joint swelling were significantly different from values for those without joint swelling. In a study of 51 RA patients by Terslev and coworkers, CF decreased and RI increased significantly 1 month after intra-articular administration of corticosteroids.

No effect on vascularization of the wrist was seen in a study by Boesen and associates of 25 RA patients 4 weeks after intra-articular corticosteroids or commencing etanercept treatment, but Strunk and colleagues described effects on two- and three-dimensional Doppler assessments of joint vascularization 7 days after intra-articular corticosteroid injection, with gray-scale signs of inflammation remaining at the follow-up examination.

The development of RA erosions can be tracked with ultrasound. In a series of studies, Backhaus and coworkers and Scheel and associates followed a cohort of RA patients with assessment of their MCP and PIP joints by repeated MRI, radiography, and ultrasound examinations. As detected with all modalities, the number of erosions increased, suggesting that ultrasound also can follow the developmental course of erosions. Ultrasound detected fewer erosions on follow-up examinations than MRI but more than radiography. Different results were reported by Hoving and colleagues for the wrist, MCP, and PIP joints during a 6-month follow-up of 46 early RA patients, possibly because of inclusion of the wrist joint in the assessment. In a study by Bajaj and coworkers, selected MCP, PIP, and MTP joints of 21 early RA patients were followed for 6 months and showed more erosive progression on ultrasound than on radiography.

Studies attempting to establish a system of holistic ultrasound assessment of RA patients are emerging. Scheel and associates proposed, based on examinations of 10 healthy persons and 46 RA patients, a 6-joint count (2nd through 4th MCP and PIP joints) for evaluation of treatment efficacy in the small joints of the hands. After ultrasound assessment of 94 RA patients, Naredo and colleagues concluded that a 12-joint count for effusion, synovitis, and power Doppler signal, including bilateral wrists, second and third MCPs, second and third PIPs of the hands, and knee joints, highly correlated with corresponding 60-joint ultrasound counts for evaluation of overall inflammatory activity.

The data strongly suggest that it is possible to follow destructive and inflammatory changes in RA with ultrasound. However, important reliability issues must be answered before the method becomes established for scientific and clinical purposes.