Summary of studies directly correlating MRI features with synovial pathobiology: MRI characteristics.
| Study | Description | Early vs Est RA | Synovial Sampling Technique/joint Sampled | Time from MRI to Synovial Sampling | Concomitant DMARD/steroid Therapy Controlled? | Joint Assessed by MRI/biopsy Site Predetermined by MRI Image? | Feature Scored | MRI Acquisition | Assessment |
|---|---|---|---|---|---|---|---|---|---|
| Konig, et al30 | Case-control cross-sectional study of 20 RA and 2 controls (OA). 8 pts with paired MRI and synovial tissue data. Main aim of study was to compare T1 and contrast-enhanced T2-weighted MRI images with histological evidence of synovitis. | Est | Arthroscopic and arthroplastic/knee | 3 weeks | No/no | Knee yes | Synovial membrane, joint capsule, hyaline cartilage, subchondral bone marrow, juxtaarticular muscle tissue, pannus | 1.5T + contrast | DCE with ROI analysis |
| Tamai, et al21 | Cross-sectional study of 9 RA. To clarify whether signal enhancement in dynamic MRI is dependent on the severity of pathologic findings in the rheumatoid synovium. | Est | Arthroplastic/knee | 1–15 days | No/intraarticular steroid for 3 mos prior to study inclusion or during interval between MRI and arthroplasty was not permitted. | Knee/yes | Synovitis | 1.5T + contrast | DCE |
| Gaffney, et al24 | Cross-sectional study of 21 RA to develop a method for quantifying acute synovial inflammation in RA using MRI. | Est | Blind needle biopsy/knee | Just prior* | No/no | Knee/no | Synovitis | 0.5T + contrast | DCE |
| Gaffney, et al25 | Cross-sectional study of 21 RA. To develop a quantitative technique for assessing synovial vascularity based upon contrast-enhanced MRI. | Est | Blind needle biopsy/knee | Just prior* | No/no | Knee/no | Synovitis | 0.5T + contrast | DCE |
| Ostergaard, et al22 | Cross-sectional study of 17 RA and 25 OA joints. To evaluate the relationship between synovial membrane and joint effusion volumes determined by MRI and macroscopic and microscopic synovial pathologic findings in pts with RA and OA. | Est | Arthroscopic or arthroplastic/knee | 1–25 days | No/no | Knee/yes | Synovial volume | 1.5T + contrast | Static (quantitative assessment) |
| Ostergaard, et al23 | Cross-sectional study of 17 RA and 25 OA joints. To evaluate dynamic as well as static Gd-enhanced MRI as measures of synovial inflammation in arthritis, by comparison with macroscopic and microscopic synovial pathology. | Est | Arthroscopic or arthroplastic/knee | 1–25 days | No/no | Knee/yes | Synovitis | 1.5T + contrast | Static (quantitative assessment) and DCE |
| Veale, et al37 | RCT of 13 RA with active resistant knee synovitis randomized to intraarticular injection of placebo, 0.4 mg or 40 mg of anti-CD4. Underwent MRI and baseline and Day 42 synovial biopsy. | Est | Arthroscopic/knee | Same day | Yes/yes, stable DMARD/steroid doses for 3 mos prior to study entry | Knee/yes | Synovitis | 1.5T + contrast | DCE |
| Ostendorf, et al26 | Cross-sectional study of 22 RA. Aim was to evaluate MRI findings in the MCP joints of pts w/RA macroscopically using miniarthroscopy. | Early (< 1.5 yr, 9 pts) + Est (13 pts) | Mini arthroscopy/MCP | 24 h | No/no | MCP/no | Synovial volume, synovial activity, effusion, joint space narrowing, bony alterations, tenosynovitis, BME | 1.5T + contrast | SQ assessment, 0–3, of each of 7 MRI variables |
| Takase, et al27 | Cross-sectional study of 10 RA and 5 OA. To simultaneously examine US, MRI, and histopathology of joint lesions in RA or OA pts who required knee joint arthroplasty. | Est | Arthroplastic/knee | 24 h | No/no | Knee/biopsy site determined by preoperative US scan | Synovitis | 1.5T + contrast | OMERACT-RAMRIS SQ, 0–3 |
| Axelsen, et al28 | Cross-sectional study of 17 RA. To determine whether DCE-MRI evaluated using semiautomatic image processing software can accurately assess synovial inflammation in RA knee joints. | Est | Arthroplastic/knee | 0–25 days | No/no | Knee/yes | Synovitis | 1.5T + contrast | DCE (semiautomatic quantification) |
| Buch, et al33 | Prospective cohort study (interventional open-label clinical trial) of 16 TNFi-resistant RA pts. MRI and synovial biopsy were performed at baseline and 16 weeks following IV ABA therapy to determine the synovial effect of ABA. | Est | Arthroscopic/knee | 0–2 days | Yes/yes, stable doses of DMARD 28 days prior to inclusion/low dose corticosteroids permitted. | Knee/no | Synovitis | 1.5T + contrast | DCE |
| Kirkham, et al 38 | Prospective cohort study of 60 pts w/RA. To explain the wide variability in joint damage progression from measures of pathologic changes in the synovial membrane. | Early (34% < 2 yrs) + Est | Arthroscopic/knee | Baseline study assessments | No/no | 2nd–5th MCP joints/no | Bone erosion | 1.5T | OMERACT-RAMRIS MRI score |
| Vordenbäumen, et al31 | Cross-sectional study of 9 pts. The objective was to analyze whether MRI synovitis relates to histological signs of synovitis in small RA joints. | NA | Arthroscopic/MCP | Up to 1 week prior | Partial all pts MTX, + 6 pts biologic/NA | MCP2/No | Synovitis | 3T + contrast | DCE |
| Vordenbäumen, et al35 | Cross-sectional study of 10 pts to analyze whether synovial markers within an MCP joint reflect global disease activity measures in RA. | NA | Arthroscopic/MCP | Up to 1 week prior | Partial, all on MTX +/adalimumab/NA | Dominant MCP/no | Synovitis/BME/erosion | 3T + contrast (6 pts), 0.2T + contrast (4 pts) | RAMRIS |
| Anandarajah, et al32 | Retrospective cohort study of 15 pts recruited to examine whether RA pts who meet remission criteria manifest inflammatory synovitis. 7 pts included with paired MRI/synovial tissue. | Est | Arthroplastic/knee, wrist, hip, elbow, thumb | 1–4 mos prior | No/no | Knee (5), wrist (5), hip (2), elbow (2), thumb (1)/no | Synovial proliferation, BME, effusion, erosion | 1.5T + contrast | SQ score, 0–3 |
| Paramarta, et al29 | Cross-sectional study in 41 pts (20 RA, 13 SpA, 8 other) aimed to compare the presence and extent of synovitis and enthesitis in early untreated SpA and RA by paired MRI and synovial histopathology. | Early (< 12 mos) | Mini-arthroscopy/knee, ankle | Not reported | Yes/yes | Knee, ankle/no | Synovitis, enthesitis | 1.5T + contrast | SQ score, 0–3 |
| Kennedy, et al34 | Prospective cohort study of 16 RA and 4 PsA. The aim was to compare the effect of TNF-blocking therapy on hypoxia in vivo, macroscopic and microscopic inflammation, and MRI variables using sequential MRI and synovial biopsy. | Est | Needle arthroscopy/knee | 24–72 hr prior | No/no | Knee/no | Synovitis | 1.5T + contrast | DCE SQ, 0–3 |
| Maijer, et al 36 | Prospective cohort study of 47 early arthritis pts (14 RA, 22 unclassified, 6 SpA, and 5 other arthritides). The aim was to examine whether DCE-MRI can be used as an objective measure of synovial inflammation using pharm. modeling. | Early (< 1 yr) | Arthroscopic | Not defined | DMARD-naive/not defined | Knee | Synovitis | 1.5T + contrast | DCE (quantitative assessment using pharm. modeling) |
↵* Exact timing not specified. MRI: magnetic resonance imaging; Est: established; RA: rheumatoid arthritis; DMARD: disease-modifying antirheumatic drug; OA: osteoarthritis; Gd: gadolinium; RCT: randomized controlled trial; MCP: metacarpophalangeal; US: ultrasound; DCE: dynamic contrast-enhanced; TNFi: tumor necrosis factor inhibitor; ABA: abatacept; SpA: spondyloarthritis; PsA: psoriatic arthritis; pts: patients; NA: not available; MTX: methotrexate; BME: bone marrow edema; ROI: region of interest; SQ: semiquantitative; OMERACT: Outcome Measures in Rheumatology; RAMRIS: Rheumatoid Arthritis Magnetic Resonance Imaging Scoring; pharm.: pharmacokinetic; IV: intravenous; .