The pathogenesis of rheumatoid arthritis in radiological studies. Part II: Imaging studies in rheumatoid arthritis.

Early diagnosis of rheumatoid arthritis followed by early initiation of treatment, prevent the destruction of joints and progression to disability in the majority of patients. A traditional X-ray fails to capture early inflammatory changes, while late changes (e.g. erosions) appear after a significant delay, once 20-30% of bone mass has been lost. Sonography and magnetic resonance imaging studies have shown that erosions are seen in the first 3 months from the appearance of symptoms in 10-26% of patients, while in 75% they are seen in the first 2 years of the disease. Power Doppler ultrasound and dynamic magnetic resonance studies allow for qualitative, semiquantitative and quantitative monitoring of the vascularization of the synovium. In addition, magnetic resonance enables assessment of the bone marrow. The ultrasonographic examination using a state-of-the-art apparatus with a high-frequency probe allows for images with great spatial resolution and for the visualization of soft tissues and bone surfaces. However, the changes seen in ultrasonography (synovial pathologies, the presence of exudate, tendons changes, cartilage and bone lesions, pathologies of tendon attachments and ligaments - enthesopathies) are not only specific for rheumatoid arthritis and occur in other rheumatic diseases. Qualitative methods are sufficient for diagnosing the disease through ultrasound or magnetic resonance imaging. Whereas semiquantitative and quantitative scales serve to monitor the disease course - efficacy of conservative treatment and qualification for radioisotope synovectomy or surgical synovectomy - and to assess treatment efficacy.

Introduction

An understanding of the pathogenesis of rheumatoid arthritis (RA) is only possible on the basis of histological studies. However, in the case of rheumatic diseases, the synovial sample is only procured if a synovectomy or joint arthroplasty is being performed. For these reasons, non-invasive imaging studies, particularly ultrasonography (US) and magnetic resonance imaging (MRI), play such a significant role in the diagnosis of these diseases, and recently also in understanding their pathogenesis. Scintigraphy or positron emission tomography are utilized in individual cases, due to their low specificity and insufficient spatial resolution(. While computer tomography, considered the method of choice for evaluating bone erosions, requires a large dose of radiation, does not allow for depicting full spectrum of possible abnormalities.

Radiograms

It is universally accepted that the radiological examination shows advanced RA changes – the narrowing of the joint space and margin erosions – that result from the destructive activity of the pannus. Erosions have been found to be a specific and relatively sensitive sign of RA, providing evidence for the radiologic examination to be included in the ACR classification of the disease in 1987(. Those diagnostic criteria were in place until 2010, when erosions were deemed a key element of RA, but often independent of synovitis. This conclusion was reached from the observation that erosions progress even once clinical and laboratory findings of joint inflammation normalize with treatment(. MR has confirmed that apart from processes in the synovium, bone damage occurs from the side of the bone marrow, which is also inflamed (look also at I part of the article). This knowledge allows us to look at characteristic RA findings differently. For example periarticular osteoporosis seen in X-ray attests to the presence of underlying hyperemia, bone marrow inflammation, which could lead to the formation of inflammatory subchondral cysts (geodes), containing inflammatory granulation tissue, and in due course also to erosions.

Radiographs are an accepted, repeatable, and accessible imaging method in RA diagnosis. They are performed in all suspected RA patients to establish the stage of the disease, with the help of semi-quantitative scales used in the set institution (for example the Larsen, Sharp, or modified Sharp scale by van der Heijde). Erosions are important in the prognosis of RA and in monitoring its treatment. Their presence in the initial X-ray, indicates a poor prognosis for the patient, due to the advanced stage of the inflammatory process and rapid progress of joint destruction. Luckily, more often patients are being diagnosed early on in the disease course. The X-ray picture is nonspecific at this point, showing edema of the periarticular soft tissue, periarticular osteoporosis(, although the latter finding is indicative of bone marrow inflammation, as previously discussed (fig. 1).

Fig. 1

X-ray of the hand: edema of soft tissues, periarticular osteoporosis, multiple subchondral inflammatory cysts in the wrist, radiocarpal bones, metacarpal bones, and phalanges, an erosion in the styloid process of the right ulna, a subchondral inflammatory cyst in the styloid process on the left side

New criteria for diagnosing RA (Modified Rome Criteria from 2010) serve as evidence of the failure of X-rays to diagnose early RA, as they do not include any imaging studies. The X-ray examination does not show early inflammatory changes, but the value of the remaining imaging methods has not been fully established, although the utility of US and MRI is emphasized.

US and MRI studies in the diagnosis of RA

US and MR are increasingly used in the diagnosis of early inflammatory changes in the course of RA. The US and MRI picture is a reflection of pathological processes taking place in the joints, while the MRI study additionally reveals pathologies in the bone marrow(. The US examination is particularly valued by rheumatologists, being accessible, relatively inexpensive and non-invasive, thus enabling frequent follow-up studies. Moreover, this imaging method does not necessitate immobility, unlike MRI for which this is a requirement; thus US is frequently used in children or patients in severe pain. Assuming that it is conducted with a modern US apparatus and a high-frequency probe, the US study can have great spatial resolution comparable to MRI, and can image soft tissues, erosions and geodes. Although unlike MRI, US does not reveal pathologies within the bone marrow, like bone marrow edema(. Spectrum of abnormalities visible in US is not RA specific and may be visible in other rheumatoid arthritis. These include:

pathologies of the synovium (thickening, hypervascularization, and fibrosis of the synovium within joint cavities, tendon sheats, and bursae);

exudates, usually coexisting with sunovium abnormalities;

tendons’ pathologies (tenosynovitis/tendovaginitis/partial/total tears);

inflammatory cysts, erosions, cartilage pathologies;

enthesopathies.

Early pathologies visualized in US are the thickening of the synovium, often coexisting with effusion, both being the result of hyperplasia of intima cells and inflammatory infiltrates within the subintima layer (fig. 2). Such an abnormal synovium could be seen lining the joint cavities, tendon sheaths and bursae. The results of the US study are usually expressed in a qualitative manner (a pathology is present or not), although several semiquantitative scales have been proposed. According to the Leeds Score(, the thickening of the synovium is evaluated in grey-scale from 0 to 3: grade 0 – lack of thickening, 1 – mild, 2 – moderate, and 3 – significant degree of thickening. Numerous articles have been published on the use of power Doppler ultrasound (PDUS) in differentiating the early inflammatory phase from the beginning of the destructive phase, in which a well-vascularized pannus is seen( (fig. 3). Other classifications take into account the results of the PD examination: grade 0 – lack of flow, 1 – one to two vessels visible within the synovium, 2 – numerous vessels up to 50% of the thickened synovium, 3 – numerous vessels occupying over 50% of the volume of the synovium. This is slightly modified in another study(: grade 0 – normal/ minimally vascularized synovium, 1 – mild degree of hyperemia, 2 – moderate degree, 3 – significant degree of synovial vascularization. There are high expectations for quantitative measurement methods. Ellegaard et al.( have proposed the CF (color fraction) index, which is the ratio of the number of color pixels to all the pixels in a selected range of interest (ROI). The authors chose ROIs to include pathological synovium in 3 standard planes of cross-sections through the radiocarpal, midcarpal, and carpometacarpal joints in a radial plane, ulnar plane, and one between them. The CF index was found to correlate with the clinical indicators of inflammation: number of swollen joints, ESR, CRP and DAS 28.

Fig. 2

The US examination: thickening of the synovium without increased vascularity in the radiocarpal and midcarpal joints

Fig. 3

The US examination: thickening and increased vascularity of the synovium in the radiocarpal and midcarpal joints

Qualitative methods may be sufficient for the disease diagnosis, whereas semiquantitative and quantitative scales serve to better monitor the disease course, including qualification to and follow-up radiofrequency synoviorthesis and, if not successful, to surgical synoviectomy. A decrease in the number of involved joints/sheaths/bursae (a reduction or complete lack of vascularization and fibrosis of the synovium) will be seen in the case of a good response to treatment( (fig. 4). A strict correlation has been shown between synovitis as assessed by PD and the number of Th17 lymphocytes in the joint fluid of RA patients. Such a finding confirms the utility of the PD study in monitoring the course of RA(.

Fig. 4

The US examination: a fibrosed synovium in the metacarpo-phalangeal joint of the 2nd digit, with erosions on the joint surfaces

As with US, the MRI picture can be evaluated using qualitative, semiquantitative and quantitative methods( (fig. 5). The choice of analysis scale depends on the clinical problem and its circumstances (for example, preliminary diagnostics, treatment monitoring, clinical trial). One of the better-known semiquantitative assessment scales of RA in the wrist and hand joints is RAMRIS, or its simplified version called SAMIS(. The RAMRIS system was developed by the EULAR-OMERACT group in 2002( and allows for the assessment of: synovitis (0 – normal picture, 1 – minor hypervascularization of the synovium, 2 – moderate degree of vascularization, 3 – significant degree of hyperemia), bone marrow edema (0 – none, 1 – involves 1–33% of the volume of the set bone, 2 – takes up 34–66% of the volume, 3 – involves 67–100% of the volume of the set bone), and erosions (graded 0 to 10, based upon the area of the erosion with respect to the volume of the set bone, in 10% intervals). Qualitative methods of evaluating the synovium rely upon measuring the thickness of the synovium, the volume of erosions, the volume of the synovium at a set surface or upon measuring the volume of the synovium in the set joint that is inflamed (inflammatory load)(. In addition, quantitative methods include measurements of contrast enhancement (IS) after intravenous administration of contrast(. Numerous studies provide evidence that the changes visualized in MRI reflect synovitis (table 1). For example, MRI-based measurements of synovium volume and its vascularity have been shown to correlate with disease activity and the number of erosions after 1 year.

Fig. 5

The MRI examination, frontal views in PD, FS before (A) and after contrast administration (B): bone marrow edema in all the carpal bones as well as the bases of the 2nd-5th midcarpal bones

Tab. 1

Scientific evidence that synovitis as seen in MRI reflects true inflammation of the synovium and that this finding is clinically and radiologically significant in RA(15)

1	Several studies have shown a strict correlation between synovitis in the knee joint, as seen in MRI, and the histopathological picture of synovitis
2	A strong correlation has been found between MRI and fine-needle arthroscopy (p < 0.01) in their assessment of histopathologically-confirmed synovitis in the MCP joints of patients with early and advanced RA
3	A high degree of compatibility (>95%) has been found between MRI and PDUS
4	The prognostic value of MRI has been confirmed with respect to the development of erosions and the clinical course of early and advanced RA
5	Several studies have shown no progression of RA (no erosion formation) in the absence of synovitis
6	One study revealed that compared to the clinical and biochemical evaluations, MRI findings of initial very advanced erosive changes and synovitis within the wrist bones were the best predictors of radiological progression of the disease in the hands, wrists and feet within a 10-year observation period

The quantitative DCE-MRI (dynamic contrast-enhanced MRI) study relies upon sequential imaging of subsequent MRI layers before and immediately after contrast administration, during circa 10–11 min. The study results in a curve of contrast flow, that reflects the time-flow of synovial enhancement. The measurements obtained from such a curve (ex. maximal enhancement, initial enhancement, time to reach enhancement) are a sensitive reflection of various physiological parameters, including the perfusion of the synovium and capillary permeability(. These measurements strictly correlate with the degree of synovial vascularization and its inflammation; hence it is assumed that DCE-MRI would be a good marker of inflammation in RA(. A crucial limitation for the use of quantitative methods is their data processing time, which can reach 1 hour. A promising solution is the automated Dynamika system(. This program, in a rapid and quantitative manner, generates colorful maps showing maximal contrast enhancement and contrast absorption in set areas (fig. 6 A). The colorful maps are then evaluated quantitatively, after selecting a range of interest (ROI) (fig. 6 B).

Fig. 6

A. Left: the mapping of the slices for the cross-sectional planes at the level of the MCP joints. Middle: selected 20 images, with magnification of the first and last images. The latter shows significant inflammation in the 3rd MCP joint. Analysis of the data is performed visually. Right: analysis using the Dynamika system: maps of maximum enhancement (top left) and Gd-DTPA uptake (bottom right). The tissues with the most inflammation are shown in bright yellow-white, while those with the least inflammation – in dark red colors. Tissues which uptake Gd are in shades of green and red. B. Curves showing the various phases of contrast flow in the studied tissues, from the left: contrast uptake, plateau, and wash-out phases

The DCE-MRI study holds much promise, although it leaves a lot of issues up for discussion, for example which sequence, layer thickness, and imaging planes should be used, or the where to set the ROI, as the synovium is histologically different in various areas of the same joint. Ideally, the analysis should take into consideration the whole volume of the enhanced synovium. The enhancement coefficient depends on the volume of the synovium occupied by the plasma, on the relaxation time of the synovium in T1 images before contrast administration, on the fraction of the synovium soaked by extravascular fluid, intracellular fluid, all of which prevent the evaluation of inflammatory intensity(. Among all of the DCE-MRI parameters analyzed, the early enhancement rate (EER) was found to best correlate with the development of erosions, ESR, disease activity as evaluated by DAS 28(. It better correlates with results of histological studies and response to treatment than the late static enhancement (SE) variable. Other researchers prefer the relative enhancement rate (RER) over the EER, as the former is independent of IS, and proportional to the gadolinium concentration in the synovium(. Theoretically, the RER is dependent to a greater degree on the vascularity of the synovium and capillary permeability than the parameters of maximum enhancement (ME) or the late, SE. For this reason, RER is expected to be the best marker of disease activity in RA. Most researchers measure RER between 30 and 60 seconds(. In addition, RER has been found to correlate with the results of FDG-PET, suggesting a relationship between enhancement and the metabolic activity of the synovium(. Another benefit to this imaging method, are the possibilities of analyzing contrast enhancement curves, particularly in the differential diagnosis of early RA. In a study of van der Leij et al.(, it was shown that the curve demonstrating a rapid enhancement (which reflects increased vascularity of the tissue) with a subsequent rapid contrast wash-out (due to the increased permeability of capillaries) allows for the differentiation of patients with early RA from healthy controls. Confirmation of these findings on a larger group of patients would provide better proof of diagnosing RA at this early stage, at which appropriate therapy could give the best treatment effects.

Conclusion

Multiple studies have shown a strict correlation between PDUS, MRI, and clinical plus histopathological parameters of the disease. Both methods enable monitoring of the RA course. From the scientific perspective, the most interesting would be to assess the usefulness of dynamic MR-imaging in the differential diagnosis of early arthrites.