Keywords knee - imaging - gout - rheumatoid arthritis - spondyloarthropathy
The knee is frequently affected by various inflammatory diseases, such as crystallopathies,
like gout, as well as autoimmune and autoinflammatory arthritides, such as rheumatoid
arthritis (RA), juvenile idiopathic arthritis (JIA), and spondyloarthropathies (SpA).
These conditions present distinct radiologic features. For instance, RA often leads
to rapid joint destruction, whereas in others, like gout, the damage to joints may
take years. Historically, the diagnosis relied on radiography, ultrasonography (US),
and magnetic resonance imaging (MRI). For crystallopathies, dual-energy computed tomography
(DECT) is now also used. Recent advances have introduced hybrid techniques, enhancing
early diagnosis, prognostication, and treatment monitoring.
Advancements in Imaging Modalities
Advancements in Imaging Modalities
Ultrasonography
Knee US has traditionally been the initial examination in the diagnostic pathway of
inflammatory arthropathies, often following a radiographic examination. Its primary
role is in the early diagnosis, monitoring of treatment, and guiding interventions.
The strength of US is its ability to evaluate soft tissues in detail including the
joint, tendons, and ligaments. Doppler interrogation allows a real-time assessment
of synovial inflammation.
Examination of the knee is usually performed with a high-frequency US transducer,
generally 5–18 MHz, that provides superior spatial resolution of superficial tissues.[1 ] Recent advances in US include high-frequency transducers of 20 to 50 MHz for the
assessment of musculoskeletal (MSK) tissues and newer technologies such as microflow
imaging and soft tissue elastography.[1 ]
Microflow imaging has been a breakthrough in the diagnosis of inflammatory arthropathies,
specifically RA[2 ] and JIA.[1 ] Its sensitivity surpasses traditional methods like power Doppler and color Doppler
because it can detect even small vessels without the need for administering contrast[2 ] ([Fig. 1 ]). The advantages of this technique include the assessment of blood flow in the microvasculature,
effective separation of flow signals from tissue motion artifacts, high resolution
of images, minimal motion artifact, and high frame rates.[2 ]
Fig. 1 A 51-year-old man with rheumatoid arthritis: knee joint effusion, synovial thickening,
and vascularization. More vessels are seen using (a ) microflow imaging than (b ) power Doppler.
Superb microflow imaging (SMI) can be performed in either standard color or even more
sensitive monochrome mode where the background signals are subtracted and only vessels,
including those with the lowest velocity flow, are seen.[2 ] A quantitative image analysis is achievable using vascular index software that allows
the operator to set a region of interest (ROI) of a different size in a two-dimensional
(2D) static image and automatically calculate the number of color pixels of vascular
signal within all pixels of the ROI.[1 ]
[2 ] SMI may significantly improve patient management at the stage of initial diagnosis,
follow-up, and remission.[1 ]
In MSK radiology, application of quantitative contrast-enhanced US is limited mainly
to the evaluation of muscle and tendon perfusion, or soft tissue tumors in the context
of research.
Elastography can assess tissue stiffness. It is particularly useful to evaluate superficial
tissues (skin, subdermis, fascia, muscles, and tendons) in juveniles and adults with
scleroderma, polymyositis, and dermatomyositis[1 ]
[2 ] ([Fig. 2 ]). The two types of elastography are strain and shear-wave elastography (SWE). The
latter is an objective quantitative technique that measures the absolute elasticity
value of soft tissues.[2 ] It uses an acoustic radiation force pulse sequence to generate shear waves that
propagates through the tissues perpendicular to the US beam, causing transient displacements.[1 ]
Fig. 2 A 6-year-old boy with localized scleroderma with skin hardening on the lateral side
of the left thigh. (a ) Superficial tissue evaluation with 24 MHz transducer shows skin thickening with
increased vascularity in superb microvascular imaging on the affected left side that
confirms active lesions. (b ) Normal skin thickness, echogenicity and vascularity of the contralateral right thigh.
(c,d ) Compression elastography confirms skin stiffness of the affected left (L) side in
comparison to the right (P) side. (e, f ) Shear-wave elastography confirms skin stiffness on the affected left (L) side, in
comparison the normal right thigh (f).
SWE images are automatically coregistered with standard B-mode images to provide quantitative
color elastograms with anatomical specificity. Shear waves propagate faster through
stiffer and inflamed tissues, as well as along the long axes of tendons and muscles.[1 ] In rheumatology, SWE has mainly been used to evaluate tendon and muscle pathology,
presenting shear waves velocity differences depending on the disease activity.[1 ]
[2 ] For example, softening of the patellar tendon in the course of arthritis was reported
in the literature.[1 ]
Magnetic Resonance Imaging
MRI enables comprehensive assessment of soft tissues, cartilage, and bone marrow.
Its high contrast resolution and multiplanar capability allows the identification
of early inflammatory changes and treatment monitoring.[3 ] Unlike US, MRI can visualize the bone marrow and provide access to all structures
of the joint.
In the last few decades, novel MRI techniques have emerged: a robust fat suppression
(FS) technique using chemical shift imaging (CSI), diffusion-weighted imaging (DWI),
advanced quantitative MRI techniques, such as T2 mapping and T1ρ mapping, dynamic
contrast-enhanced MRI (DCE-MRI), diffusion tensor imaging, and whole-body MRI (WB-MRI).[4 ]
[5 ]
[6 ]
Robust FS techniques predominantly involve Dixon sequences that use CSI. They take
advantage of the slight difference in resonance frequency between water and fat protons
to provide a series of four sets of images: in-phase (IP), out-of-phase (OP), water
only, and fat only.[4 ] Although Dixon described it in the 1980s, it has only recently become possible to
associate this method with spin-echo (SE)–based sequences, the backbone of MSK MRI
protocols, opening the door for multiple applications in this field. CSI MRI can also
be used to probe bone marrow fat content quantitatively, by measuring the signal drop
between IP and OP images or by calculating the fat fraction. This can be done both
on gradient-echo (GRE) and SE sequences.[4 ]
The Dixon method has proven to be more robust to magnetic field inhomogeneities than
chemical shift selective suppression while providing a better signal-to-noise ratio
than short tau inversion recovery sequences.[4 ] Therefore, it is an appealing technique for large field-of-view (FOV) imaging of
the bone marrow, whereas in knee assessment its use is generally reserved to assess
tumors rather than inflammatory conditions. The robust FS and reduced examination
time are additional advantages of Dixon acquisitions.
DWI is based on the diffusion of water molecules in tissues and obtained by two symmetrical
gradient pulses on conventional MR sequences, the first causing the phase shift of
water molecules and the second one canceling it[6 ] ([Fig. 3 ]). It has not gained popularity, despite reports of comparable accuracy in diagnosing
synovitis and predicting radiologic and clinical response in patients with JIA.[6 ]
[7 ]
[8 ] Some challenges associated with this technique include long study time, high sensitivity
to field inhomogeneities, the need for strong gradients, and subjectivity in positioning
the ROIs that hampers the objective quantification of disease.[6 ] Nevertheless, some potential solutions to resolve this shortcoming have already
been proposed.[6 ]
Fig. 3 Axial RESOLVE Diffusion Weighted Imaging (TR 4710, TE 71, number of averages 2, slice
thickness 3.5 mm) of the knee demonstrating the thin, irregular and inhomogeneous
cartilage of the patellofemoral compartment (white arrows on the b50 s/mm2 and b800 s/mm2 images, and on the apparent diffusion coefficient map respectively in a, b, and c).
Intravoxel incoherent motion (IVIM) is a method developed by Le Bihan and colleagues
in 1988 to evaluate the microscopic motion of water molecules that simultaneously
quantifies diffusion and microperfusion using diffusion-weighted sequences.[6 ] Since then, IVIM has been applied in various fields including rheumatology, with
very good results.[6 ]
Quantitative assessment of synovial vascularization on DCE-MRI can evaluate the degree
of synovial inflammation.[9 ] DCE-MRI consists of assessing tissue perfusion through serial acquisitions of images
before and after a bolus of intravenous contrast injection and the review of the variation
of MR signal intensity of the tissues of interest, both qualitatively and quantitatively,[4 ] using T1-weighted GRE sequences that allow a high temporal resolution (1–10 s).[6 ] These generate signal intensity curves based on the distribution of contrast medium
in the intra- and extravascular spaces, from which quantitative perfusion parameters
can be extracted by applying a pharmacokinetic or a heuristic approach.[6 ]
The important role of DCE-MRI for assessing synovitis in the knee joint is well established,
especially in patients with early disease not diagnosed by conventional MRI.[6 ] DCE-MRI may even eventually replace the semiquantitative Rheumatoid Arthritis Magnetic
Resonance Imaging Score (RAMRIS).[6 ] Encouraging results in this direction emerged from an exploratory study about the
effect of tofacitinib, in which both DCE-MRI with heuristic analysis and a novel quantitative
adaptation of the RAMRIS, based on active appearance modeling, called Rheumatoid Arthritis
Magnetic Resonance Imaging Quantification (RAMRIQ), were applied, showing an early
response to treatment.[6 ]
Several new techniques are used for cartilage assessment ([Fig. 4 ]). T1ρ is a technique sensitive to low-frequency interactions of macromolecular protons
and bulk water.[6 ] T1ρ relaxation time describes spin-lattice relaxation in the rotation frame in the
presence of an external radiofrequency pulse in the transverse plane. Images acquired
with different spin-lock frequencies are used for the calculation of T1ρ in each pixel
by fitting a signal to exponential function.[6 ]
Fig. 4 Axial Multiple Echo Data Image Combination (MEDIC), T2*-weighted spoiled gradient-echo
sequence of the knee nicely showing (a ) patellofemoral cartilage with regular thickness and signal intensity in a 47-year-old
male patient. (b ) Cartilage loss with irregular margins (white arrows) in a 70-year-old woman with
severe osteoarthritis.
Similarly, in T2 mapping, which maps the transversal relaxation constant, T2 values
are calculated from a multiecho multislice sequence by fitting a signal with the exponential
curve.[6 ] In patients with RA, T1ρ and T2 values of the knee cartilage were higher than in
healthy volunteers except for the patella and lateral tibial plateau.[6 ] Similar results using T1ρ were achieved on tissue samples of 5 and 14 patients,
respectively, with RA and osteoarthritis (OA) who underwent total knee arthroplasty.[6 ] The authors demonstrated that T1ρ is useful to detect and map early stages of cartilage
degradation in both diseases and that changes due to RA are associated with higher
T1ρ values.[6 ]
Delayed gadolinium-enhanced MRI of cartilage (dGEMRIC) allows the indirect quantification
of glycosaminoglycans after the injection of gadolinium diethylenetriamine-penta-acetic
acid (Gd-DPTA[2 ]) in areas affected by cartilage loss. Typically, low dGEMRIC indexes are seen in
RA.[6 ] The ultrashort echo time and the zero echo time MRI sequences in the knee allow
the detection of rapidly decaying transverse magnetization in the so-called short
T2 tissues such as cartilage, fibrocartilage (menisci), cortical bone, and tendons.[6 ]
Three-dimensional (3D) isotropic MRI acquisitions have several advantages over 2D
MRI[10 ] ([Fig. 5 ]). 3D MRI provides improved spatial resolution and can generate high-quality reformats
from the original data set.[10 ] Thin continuous slices improve through-plane resolution that reduces artifacts from
partial-volume averaging, can delineate erosive changes and synovitis better than
2D images,[10 ] and enhances the ability to delineate more subtle pathologies that might be undetectable
with conventional 2D MRI.[10 ] The 3D isotropic MRI pulse sequences allow reformatting of the high-quality images
in any desired plane to improve image assessment,[10 ] whereas the ability to reformat images from a single data set can reduce the number
of sequences and decrease scan time.[10 ]
Fig. 5 Isotropic knee images, proton-density fat suppression sequence with 0.8-mm slice
thickness, and 0.4-mm spacing, TR: 1,000 ms, TE: 50 ms, flip angle: 90 degrees. (a ) Sagittal image, which is the original scan plane. (b ) Coronal image, reformatted from the sagittal plane data set. (c ) Axial image, reformatted from the sagittal plane data set.
The 2D and 3D GRE sequences with different types of contrast weighting can be broadly
classified into two categories: dark fluid signal sequences and bright fluid signal
sequences[11 ] ([Fig. 4 ]). Research on bright and dark fluid GRE sequences in knee MRI has focused mainly
on cartilage evaluation because there is typically good contrast between fluid, articular
cartilage, and bone.[11 ] High-resolution isotropic dark fluid GRE morphological sequences have often been
considered the standard for cartilage morphology assessment in the research setting,[11 ] but they can have long scan times, forcing a compromise of resolution or quality
when optimized for clinically feasible acquisition times. Bright fluid high-resolution
3D GRE sequences, such as the 3D double-echo steady-state sequence, offer faster acquisition
times and favorable fluid-to-cartilage contrast, resulting in improved assessment
of the cartilage surface.[11 ]
Dual-Energy Computed Tomography
Computed tomography (CT) with its multiplanar capability and excellent depiction of
bone is the best method to assess erosions.[12 ] However, due to radiation concerns, it is infrequently used solely for this purpose.
Spectral CT operates on the principle that tissue attenuation is influenced not just
by density but also by the atomic number (Z) and the photon beam's energy.[4 ] By leveraging these properties, spectral CT can characterize and quantify specific
tissue components.[4 ]
The most prevalent subset of spectral CT is DECT, using two distinct X-ray energy
spectra.[4 ] In dual-source DECT, two tube-detector pairs are utilized, with adjustable tube
voltages, facilitating fast single energy combinations. The two separate tubes are
offset by ∼ 90 degrees to each other, contributing to the material decomposition performed
only on the image domain due to the spatial offset between acquisitions.[13 ]
One technique for single-source DECT is rapid kilovoltage switching. In those scanners,
there is virtually no temporal mismatch and full FOVs with the X-ray tube switching
between 80 and 140 kVp in < 0.2 ms.[13 ] Contrary to dual-source DECT systems, the resulting projection-space decomposition
offers greater flexibility in the types of materials that can be used for data and
provides a significant advantage in minimizing beam-hardening artifacts.[13 ]
Other kinds of DECT techniques include dual-layer technology, sequential acquisitions,
or split-filter approaches. Recent innovations have introduced photon-counting detectors,
groundbreaking technology that enables multi-energy imaging, through direct counting
of individual incoming photons and measurement of their energy level.[4 ] These detectors offer superior spatial resolution, dose reduction, and enhanced
material differentiation, and they are less susceptible to beam-hardening artifacts.[4 ]
[13 ]
These special CT techniques offer material characterization based on two or more X-ray
photon energy-dependent attenuations.[6 ] The distinct absorption properties of calcium and urate have positioned DECT as
the leading technique for gout investigation, even in its early stages.[6 ]
[14 ] This allows for both qualitative and quantitative determination of monosodium urate
(MSU) deposition in joints and tissues and resulted in high sensitivity (88%) and
specificity (90%) in a meta-analysis.[6 ]
[15 ] After the acquisition scan, with a protocol optimized for detecting small urate
deposits, the data sets allow for the characterization and quantification of different
materials and structures, such as urate deposits and soft tissues.[6 ] This is achieved by color-coding features on 3D multiplanar volumetric images. DECT
excels in visualizing deeper or intricate structures and in displaying the anatomical
extent of gouty deposits.[13 ] Artifacts produced during DECT scanning and postprocessing can lead to false-positive
results. The common artifacts encountered in DECT are well known. [13 ]
Pseudogout, another acute inflammatory monoarticular crystal arthropathy characterized
by the deposition of calcium pyrophosphate dihydrate crystals, can present similarly
to gout on radiographs, US, or CT. DECT can diagnose pseudogout by revealing the presence
of calcium and the absence of MSU crystals.[16 ] The sensitivity of DECT in detecting pseudogout is thus lower than that for gout
detection because it is a diagnosis of exclusion.[16 ]
[17 ]
Other application of DECT around the knee include bone marrow imaging.[16 ] By using virtual non-calcium (VNCa) reconstructions, DECT can detect nontraumatic
bone marrow edema (BME) of the hip and knee with high sensitivity and specificity.[4 ] Therefore, if MRI is contraindicated, DECT can help depict BME associated with cortical
erosions, confirming osteomyelitis. In inflammatory arthropathies, VNCa images can
evaluate osteitis in the form of BME, correlating well with MRI.[16 ]
Finally, regarding metals, DECT is a validated technique for metal artifact reduction
and also used to diagnose iron-containing entities. It can characterize and color-code
iron, which can be used to diagnose synovial giant cell tumors This benign neoplastic
disorder is characterized by hemosiderin deposition, along with mononuclear cells
and multinuclear giant cells. In patients with soft tissue masses around the joint,
the presence of iron at DECT is indicative of this disease.[16 ] DECT can also be used in metallosis to detect metal debris and pseudotumors.[16 ]
Nuclear Medicine and Molecular Imaging
Nuclear imaging relies on the intravenous injection of radiopharmaceuticals to assess
the distribution of hematopoietic or reticuloendothelial cells.[4 ] Imaging of the hematopoietic component can be achieved by white blood cell (WBC)
scintigraphy, by injecting the patient's WBCs after they have been radiolabeled in
vitro with technetium (Tc)-99m or indium-111, or by injecting Tc-99m–labeled mouse
anti-granulocytes monoclonal antibodies or antibody fragments.[4 ] The most common indication for WBC imaging in the knee is to assess a periprosthetic
infection and to differentiate it from mechanical loosening.
Bone scintigraphy with intravenous injection of Tc-99m–labeled diphosphonates allows
imaging of osteoblastic activity.[4 ] In rheumatology, they are used to diagnose insufficiency fractures, although the
knee is not a predilection area for that pathology.
Radiosynovectomy (RSV) is a treatment that destroys the hypertrophic synovial membrane
using ionizing radiation. It requires the application of β-emitting radionuclides
to treat the chronic inflammation of the joints.[18 ] For the knee and other large joints, the procedure is performed using an intra-articular
injection of yttrium-90 silicate/citrate. RSV is an overall safe technique with known
contraindications.[18 ] Despite the use of ionizing radiation, overall exposure is low, and no evidence
indicates an increase in cancer risk when compared with the general population in
adult patients.[18 ]
Hybrid Imaging
Single-photon emission computed tomography (SPECT), positron emission tomography (PET)/CT,
and PET/MRI provide functional and improved 3D anatomical depiction of inflammatory
lesions, enabling semiquantitative evaluation of uptake and metabolism of radiotracers
(in SPECT) or isotopes of glucose in inflammatory cells (PET). PET/MRI provides the
simultaneous acquisition of metabolic information combined with the intrinsic high
soft tissue contrast of MRI and the lower radiation dose, which has expanded its use
in the MSK field.[6 ] PET/MRI offers several advantages over PET/CT, including a reduced radiation burden
for the patient, superior soft tissue contrast, and better coregistration of PET data
with MRI-based motion correction.[19 ]
The benefits of PET/CT have been widely described in rheumatology.[6 ] This hybrid technique allows not only a visual assessment of the affected areas
but also a (semi)quantitative evaluation of the inflammatory process, using parameters
like SUVs (i.e., SUVmax and SUVmean), metabolic active volume, and total lesion glycolysis.[6 ] In terms of radioactive labels, fluorine-18 fluorodeoxyglucose (18F-FDG) and 18F-sodium
fluoride (18F-NaF) are the most commonly applied tracers in clinical practice for
inflammatory arthritis.[6 ]
FDG-PET/CT has been proposed as an objective noninvasive adjunct to the clinical assessment
of patients with RA, contributing to both the initial diagnosis and the response to
treatment.[19 ] PET has also been used as a tool to assess progressive joint destruction in patients
with RA. The degree of hypermetabolism in joints with active destruction is higher
than that in joints with only nondestructive inflammation. Baseline or pretreatment
SUVmax is a significant predictive factor of large joint destruction at 2 years.[19 ] Follow-up FDG-PET/CT was also shown to be an effective tool in monitoring the response
to treatment of RA and other arthropathies.[19 ]
Although not commonly affecting knees, in polymyalgia rheumatica (PMR) FDG-PET/CT
was found useful to differentiate this entity from the onset of RA in older adults.[19 ] FDG-PET/CT has also been shown to be beneficial for monitoring response to treatment
in patients with PMR.
FDG-PET has shown promise as an adjunct to clinical assessment in spondyloarthropathies.
In a study of seven patients with ankylosing spondylitis, three patients with psoriatic
arthritis, and one patient with nonspecific SpA, FDG-PET accurately delineated the
inflammatory activity at both articular and extra-articular sites.[19 ] The uptake in the affected joints in these patients with SpA was nonsymmetrical
and heterogeneous (compared with the relatively symmetrical findings in RA). There
was intense tendinous, entheseal, and muscular uptake at symptomatic joints.[19 ]
Overview of Pathologic Findings in Arthropathies
Overview of Pathologic Findings in Arthropathies
Inflammatory rheumatic diseases affecting the knee joint predominantly exhibit synovitis,
frequently accompanied by effusion, that will lead to joint damage. Other tissues
affected by inflammatory conditions around the knee include bursae, tendons, entheses,
bone marrow, fat tissue, muscles, fasciae, and superficial tissues, like skin and
the subdermis.
Gout and other inflammatory arthropathies can lead to affliction of the knee extensor
apparatus, notably of the deep infrapatellar bursa and the gastrocnemius bursa. The
same entities can induce tenosynovitis and tendinopathy, as well as enthesopathy,
with gout showing a predilection for impacting the extensor side of the knee[20 ] and SpA in general showing predilection to lower limbs entheses, causing enthesitis,
a hallmark of SpA in the knee.
The concept of the “enthesis organ” suggests that tissues adjacent to the enthesis,
such as fatty tissue, bursa, and bone marrow in the bony part of an enthesis, may
also experience inflammation.[21 ] However, degenerative enthesopathic findings, especially in the lower limb, are
more prevalent, requiring a cautious diagnosis to avoid misattributing inflammatory
arthritis based solely on imaging findings.[22 ] The criteria to differentiate inflammation-related enthesitis from metabolic, age,
and overload-driven enthesopathy remain to be solidly established.[22 ]
[23 ]
In the popliteal fossa, some tendons (most often the semimembranous) may be involved
by different processes, including tenosynovitis, tendinopathy, bursitis, and enthesitis.
In the course of most of the presented diseases, the intra-articular and periarticular
fat tissue may also look abnormal on US and MRI because it can become a site of inflammatory
infiltration, in addition to synovium and subchondral bone marrow, with cells pivotal
in joint destruction.[24 ]
Gout and inflammatory arthropathies can lead to knee damage, then erosions (marginal,
later subchondral), articular cartilage damage (resulting in osteochondral lesions
and secondary OA), and knee malalignment, as well as proliferative changes, like enthesophytes
and periosteal ossifications (characteristic of SpA).
Whereas US allows intra- and extra-soft tissue assessment (except for deep intra-articular
structures), MRI is optimal for imaging cartilage and may also reveal BME, signaling
cellular infiltration, considered a pre-erosive condition and a response-to-treatment
biomarker.
MRI also offers a superior assessment of the menisci and reveals frequent infiltration
and damage by synovium in RA, whereas JIA patients often present with meniscal hypoplasia.[25 ]
Osteophytes and subchondral sclerosis stand out as distinctive features of OA that
may manifest secondarily to inflammatory arthritis.
Additionally, rheumatic diseases, such as scleroderma, dermatomyositis, and polymyositis,
exhibit a propensity to involve the knee, particularly affecting soft tissues (skin,
subdermis, fascia, muscles) ([Fig. 2 ]).
Rheumatoid nodules, emerging under the skin, are observed in up to 20% of RA patients,
predominantly appearing in areas subjected to trauma or friction.
Specific Forms of Arthropathies
Specific Forms of Arthropathies
Although the individual diseases and arthropathies described here can have multifactorial
etiologies with complex and overlapping pathogeneses, the specific forms of arthropathies
can be broadly classified ([Table 1 ]). In this review, we focus on gout and RAs, which after septic arthritis are the
most important arthropathies to exclude in the knee joint. Brief overviews of other
connective tissue diseases (JIA, Still's disease, scleroderma, SpA, and chronic nonbacterial
osteitis) are also provided.
Table 1
Classification of arthropathies
Crystal:
• Gout
• CPPD
○ Pyrophosphate arthropathy
○ Pseudogout
• Other crystals (e.g., HADD)
Inflammatory:
• Seropositive
○ Rheumatoid arthritis
• Seronegative
○ Psoriatic arthritis
○ Ankylosing spondylitis
○ Enteropathic arthritis
○ Reactive arthritis
• Pediatric: Juvenile idiopathic arthritis
Infectious
Degenerative: Osteoarthritis
Connective tissue related and other miscellaneous arthropathies:
• Systemic lupus erythematosus
• Scleroderma
• Adult-onset Still's disease
• Vasculitis
• Sjögren's syndrome
• CNO; also known as SAPHO and CRMO
Abbreviations: CNO, chronic nonbacterial osteomyelitis; CPPD, calcium pyrophosphate
deposition disease; CRMO, chronic recurrent multifocal osteomyelitis; HADD, hydroxyapatite
deposition disease; SAPHO, synovitis, acne, pustulosis, hyperostosis, osteitis.
Gout
Gout is the most common form of inflammatory arthropathy resulting from MSU crystal
deposition, and its prevalence is increasing in Western societies.[13 ]
[26 ]
[27 ] The knee, following the metatarsophalangeal joints, is the second most common location
affected by gout.[26 ] If left untreated, the persistent deposition of MSU crystals in the joints and surrounding
soft tissues can lead to progressive joint destruction.[13 ]
Diagnosing gout hinges on identifying negatively birefringent MSU crystals in joint
fluid or tophi through polarized microscopy.[13 ]
[26 ] However, this method is infrequently used in clinical practice due to its invasiveness
and other limitations.[13 ] Imaging modalities that can facilitate the diagnosis of gout and are included in
the American College of Rheumatology/European League Against Rheumatism (ACR/EULAR)
criteria: plain radiography (presence of typical bone erosions), US (presence of a
“double-contour” sign), and DECT (color-coded MSU crystals)[27 ] ([Fig. 6 ]).
Fig. 6 Patient evaluation for total knee arthroplasty. Erosion (black arrows) and tophus
(white arrowheads) is better seen on ultrasonography (US) (b ), computed tomography (CT) (g ), and magnetic resonance imaging (MRI) proton-density fat-suppressed (FS) sequence
(d ) compared with radiography (a ). Tophus is highlighted in dual-energy computed tomography (DECT) (c ) and shows massive vascularization in contrast-enhanced T1 FS MRI (h ). Sagittal US of the trochlea shows a typical double-contour sign (white arrowheads)
(f). Note that small tophi in the patellar tendon are very well visualized by DECT
and CT but not with MRI (open arrow). (e ) Radiography of the forefoot shows typical gout lesions at the first metatarsophalangeal
joint, illustrating the strength of radiography to image multiple joints with minimal
effort.
Characteristic radiographic findings are present only in the chronic stage, and therefore
US is a common modality used to assess gout. Features of early and chronic gout on
US are well described in the literature.[26 ] In early gout, joint effusions with or without hyperechoic MSU crystal foci may
be seen that measure < 1 mm (“starry sky” sign). Larger MSU aggregates, referred to
as micro-tophi, may result in a “snowstorm” appearance. The presence of MSU crystals
embedded in hypertrophied synovium may increase the specificity.[26 ]
The double-contour sign results from deposition of gout crystals on the surface of
the articular cartilage and results in a continuous hyperechoic line. It needs careful
distinction from the “cartilage interface” sign that is a hyperechoic reflection at
the cartilage border that appears only at a 180-degree angle. In chronic gout, the
tophi appear as well circumscribed, hyperechoic, or hypoechoic nodules that initially
are uniform.[26 ]
[28 ] With chronicity, they become nonhomogeneous (“wet clumps of sugar” sign), with characteristic
tiny internal hyperechoic echoes or aggregates. Juxta-articular erosions may also
be seen.
The clinical use of DECT for the diagnosis of gout diagnosis has dramatically increased
over the past decade, and its validity as a tool for the diagnosis has been established.[13 ] Based on reliable published data and its diagnostic accuracy,[13 ] DECT was incorporated into the 2015 ACR/EULAR gout classification criteria.
Sotniczuk et al assessed the diagnostic value of DECT in patients with a clinical
suspicion of gout and compared the ACR/EULAR criteria with and without DECT diagnosis.[27 ] Considering artifacts, the combination of DECT and clinical findings resulted in
the best diagnostic accuracy. Gamala et al conducted a meta-analysis of 10 studies
on the diagnostic accuracy of DECT and revealed that the pooled sensitivity and specificity
of DECT are 81% and 91%, respectively, whereas the pooled sensitivity was 55% in patients
with recent-onset disease (≤ 6 weeks).[29 ]
Other studies report sensitivities between 12% and 90%.[13 ] A systematic review in 2022 showed that DECT images are reliably interpreted with
intra-rater intraclass correlation coefficients ranging from 0.86 to 1.00.[30 ] DECT overall had very good sensitivity and specificity in established gout when
compared with joint aspiration, with ranges from 0.78 to 0.89 and 0.84 to 1.00, respectively.
DECT also performed very well against clinical criteria, with a pooled sensitivity
and specificity of 0.81 (95% confidence interval [CI], 0.77–0.86) and 0.91 (95% CI,
0.85–0.95).[30 ]
A study comparing the diagnostic accuracies of dual-source DECT and US in patients
with different gouty disease durations reported that the sensitivities of DECT for
gout within 1 year, 1 to 3 years, and > 3 years are 26.6%, 66.6%, and 90%, respectively.[31 ] They also revealed that the sensitivity of US is remarkably higher than that of
dual-source DECT in early gout and suggested US as the first choice to diagnose early-stage
gout.[13 ] Another research group reported a higher sensitivity of 80% using dual-source DECT
in 20 gout patients with symptom duration < 6 weeks.[13 ] The outcomes demonstrated that disease duration strongly affects the diagnostic
accuracy of single-source DECT, and DECT has limited diagnostic value in early-stage
gout.
These results are attributable to the limitations of DECT in spatial (0.5–2 mm) and
contrast resolution. Studies reported a concentration threshold of detectable MSU
crystals between 20% and 35%.[13 ]
[26 ]
[32 ] In clinical practice, the concentration and detectability of MSU tophi is influenced
by disease duration because it influences crystal volume and the active chemotaxis
and phagocytosis of leukocytes.[13 ] Also, postcontrast intrinsic vascularization of tophi can increase its density,
thus helping make low-concentration tophi visible.[33 ]
DECT may be used to stage patients with gout, especially in established disease, because
50% of patients have abnormal DECT scans with normal serum urate levels and no clinical
tophi.[30 ]
Other applications of DECT include prediction of disease flares at 6 months and 2
years and excess cardiovascular and all-cause mortality.[30 ]
Rheumatoid Arthritis
RA is a systemic autoimmune disease that predominantly affects the MSK system; the
knee is one of the most commonly involved joints. It induces polyarthritis and inflammation
in tendon sheaths and bursae. Chronic synovitis and osteitis contribute to the progressive
destruction of hyaline cartilage, bone, and soft tissue.[34 ]
[35 ]
Early inflammatory changes, such as synovitis, tenosynovitis, bursitis, and osteitis
(BME), can be observed on US and MRI, along with periarticular demineralization potentially
progressing to generalized bone loss that will be seen on radiographs. Although enthesitis
is not typical for RA, secondary involvement of the patellar tendon enthesis due to
deep infrapatellar bursitis is common.
Late imaging changes, indicative of joint damage, include inflammatory cysts, joint
erosions, uniform narrowing of the joint space following hyaline cartilage damage,
knee deformities, and secondary OA ([Fig. 7 ]). Bone damage (cysts, erosions, hyaline cartilage loss) is initiated by the synovitis
or osteitis or both through the outside-in and inside-out mechanism.[36 ]
Fig. 7 Knee magnetic resonance imaging (MRI) and ultrasonography (US) of a 47-year-old female
patient with rheumatoid arthritis. MRI images: (a ) coronal proton-density (PD) image with fat suppression (FS). (b ) Coronal T1. (c ) Axial proton-density FS. (e, f ) Knee US. On MRI images, synovitis (white arrowheads) and bone marrow edema (white
arrows), erosions (black arrowhead), and deep cartilage loss (black arrows) are seen.
On US, effusion with (d ) synovial thickening in the suprapatellar recess with vascularization is seen in
(e), the medial compartment (open white arrows).
Kondo et al explored the significant association between knee joint US parameters
and synovial inflammatory factors in RA, involving 44 patients, both treated (n = 25) and untreated (n = 19).[37 ] US images were quantitatively analyzed using gray-scale assessment of synovial hypertrophy
and echogenicity and power Doppler for vascularity. The study found significant correlations
between US synovial hypertrophy, power Doppler vascularity, and synovial fluid inflammatory
cytokine levels in untreated patients.
Meng et al conducted a prospective analysis of 26 knee joints from 22 patients undergoing
total knee arthroplasty for RA treatment, revealing that characteristic MRI features
of advanced knee joint RA are damage to cartilage and menisci, attributed to synovium
infiltration allowing inflammatory cell infiltration.[38 ] The most severe pathologic changes in cartilage were fibrosis, thinning, and destruction;
in the menisci, fibrosis and engulfing calcified debris were most prominent. Chaplin
observed areas of completely destroyed cartilage, often filled with soft granulation
tissue, and significantly reduced or vanished menisci in 50 patients undergoing synovectomy
for RA.[39 ] Kimura and Vainio also noted that the knee menisci underwent a more rapid degeneration
process than the cartilage in 47 patients undergoing synovectomy for RA treatment.[40 ]
Sarcopenia, a prevalent extra-articular manifestation in RA patients, is currently
defined through developing diagnostic criteria, with imaging techniques central to
measuring or estimating muscle mass quality.[41 ] According to the European Working Group on Sarcopenia in Older People 2 consensus,
dual-energy X-ray absorptiometry is the gold standard for confirming sarcopenia diagnosis;
US is recognized for its accessibility, low cost, and ease of use.[41 ] Despite potential variables affecting US imaging, parameters like cross-sectional
area (CSA) and volume measured using panoramic images offer reliability comparable
with MRI.[41 ] Salaffi et al found that MRI-CSA-25 can differentiate between sarcopenic and nonsarcopenic
RA patients, serving as an imaging biomarker of this condition.[41 ]
Juvenile Idiopathic Arthritis
The knee is the most commonly involved joint in JIA.[22 ]
[42 ] Contrary to adult arthropathies, joint damage in most children appears late; instead,
developmental disorders may occur and are typical for JIA.
Previous studies showed that synovial inflammation and joint effusion in JIA are closely
related to clinical arthritis activity.[42 ] The study on BME as an important manifestation of early bone involvement in JIA
is presently lacking.[42 ] Yang et al found a positive correlation between BME and synovial hypertrophy.[42 ] Older children and children with long disease duration had a higher risk for BME,
which was commonly a late presentation and more likely involved the weight-bearing
surfaces of the joint.[42 ] BME was accompanied by synovitis in all joints. There was no BME finding without
synovitis in this study, and there was a positive relationship between them.[42 ]
This finding is consistent with the traditional pathogenesis hypothesis for JIA that
suggests inflammation from synovitis results in cartilage damage, bone erosion, and
eventually BME. The median duration of disease in the BME group was 9 months, significantly
longer than that in the non-BME group (4 months), suggesting that BME may not happen
during the early JIA development but rather occurs gradually with the progression
of JIA. This differs from adult RA where BME was reported more frequently in early
disease development[42 ] ([Fig. 8 ]).
Fig. 8 Knee magnetic resonance imaging of a 17-year-old girl with juvenile idiopathic arthritis
(JIA). Proton-density with fat suppression images in (a ) sagittal and (b ) coronal planes, (c ) T1 in coronal, and (d ) T2 in sagittal planes. Effusion and synovitis (white arrows), subchondral bone marrow
edema of the lateral femoral loaded part of the articular surface (white arrowheads),
marginal erosions laterally (black arrow), grade 3 chondromalacia (open white arrow),
menisci degeneration, with protrusion, lateral meniscus with a bucket-handle tear
(b, c), dysmorphy of the joint secondary to JIA with hypoplasia of the femoral and
tibial condyles and articular surfaces (b, c) and ankylosis in the tibiofibular joint
(open black arrow).
Axial Spondyloarthritis
The most common form of axial SpA is ankylosis spondylitis (AS). In 10 to 20% of cases,
the disease begins with asymmetric involvement of peripheral joints, predominantly
affecting the knee and hip. Unlike RA, erosions are atypical and seldom observed in
most cases. Instead, periarticular soft tissue thickening, bone loss, uniform joint
space narrowing, osteophytes, and isolated cysts, with enthesopathy a hallmark feature,
are found. Other characteristic imaging findings for axial SpA include tenosynovitis,
bursitis, and BME[43 ]
[44 ]
[45 ] ([Fig. 9 ]).
Fig. 9 A 35-year-old female patient with axial spondyloarthritis and knee pain. (a, b ) Radiography shows mild effusion (white arrowheads). (c ) Proton-density fat-suppressed magnetic resonance imaging (MRI) shows mild and diffuse
bone marrow edema (white arrowheads). (d ) After contrast administration, T1 fat-suppressed sagittal MRI depicts synovitis
(black arrowheads). MRI of the sacroiliac joints with (e ) T1 and (f ) T2 fat-suppressed sequences shows active sacroiliitis (black arrows).
In psoriatic arthritis, enthesopathy and periostitis may be seen in the knee. BME
in the peripheral joints may be more diffuse than in other forms of SpA,[46 ] but this finding has not been assessed in the knee ([Fig. 10 ]).
Fig. 10 A 17-year-old female patient with psoriatic arthritis. (a ) Initial radiography is unremarkable. However, (b ) proton-density fat-suppressed magnetic resonance imaging shows bone marrow edema
(white arrow). (c ) It was also visualized after contrast enhancement in addition to pronounced soft
tissue inflammation (white arrowheads). Follow-up (d ) radiography and (e ) computed tomography reveal new bone formation 1 year after symptom onset.
The knee and ankle joints are commonly involved in reactive and inflammatory bowel
diseases associated with (enteropathic) arthritis.[47 ] The lesions include soft tissue thickening, periarticular bone loss, cysts, erosions,
and periostitis/new bone formation. In reactive SpA, enthesopathic lesions are common,
whereas in enteropathic SpA, imaging findings are similar to AS, described earlier.
Scleroderma, Dermatomyositis, and Polymyositis
The diagnosis of inflammatory lesions in scleroderma in the skin, subcutaneous tissues,
as well as in fascia and muscles (in deep scleroderma), is often performed by US.
The last few years have seen a spectacular development in US, with the introduction
of high-frequency probes, options for assessing microvascularity, and the use of SWE
([Fig. 2 ]). These techniques enable early diagnosis, characterization, and quantification
of disease activity and treatment monitoring[48 ]
[49 ]
[50 ]
[51 ] ([Fig. 2 ]).
In the limited form as CREST syndrome (calcinosis, Raynaud's phenomenon, esophageal
dysmotility, sclerodactyly, and telangiectasias) and in other forms like calcinosis
universalis, calcifications of the subcutaneous tissue at certain points of mechanical
stress may be best evaluated with radiography or CT.[52 ]
[53 ]
[54 ] MRI is the method of choice for assessing inflammatory changes in muscle and fascia
in the course of dermatomyositis and polymyositis in adults and juveniles, showing
“edema-like” signal intensity in affected muscle, fascia, and subcutaneous tissue
that correlates with disease activity[48 ]
[50 ] ([Fig. 11 ]). MRI can help determine the biopsy site. In chronic stages, fatty degeneration,
muscle atrophy, and calcifications may be seen.
Fig. 11 Whole-body magnetic resonance imaging in a 6-year-old girl with juvenile dermatomyositis,
T2 turbo inversion recovery magnitude images in coronal planes. (a, b ) Initial exam and (c, d ) follow-up exam after 6 months of treatment. Initially seen high signal in numerous
involved muscles of the neck, shoulder and pelvic girdles, and upper and lower limbs,
including short muscles of the feet, undergo complete resolution after successful
treatment.
Adult-onset Still's Disease
Adult-onset Still's disease (AOSD) is a rare systemic inflammatory disease of the
connective tissue that usually affects the knee and wrist.[43 ]
[48 ] Its etiology is unknown. The imaging features in the knee may resemble RA and include
periarticular soft tissue thickening, periarticular demineralization, erosions, joint
space narrowing, and early formation of bony ankylosis seen on radiography, with joint
effusion, synovitis, or tenosynovitis seen on US and on MRI. The characteristic findings
of AOSD on PET/CT are increased 18F-FDG accumulation in the spleen, bone marrow, and
lymph nodes.[55 ]
Pediatric and Adult Chronic Nonbacterial Osteitis
A new and as yet unpublished consensus introduces “chronic nonbacterial osteitis”
(CNO) as the preferred term for a syndrome previously known as synovitis, acne, pustulosis,
hyperostosis, and osteitis (SAPHO), given that many patients do not exhibit all or
even most of the eponymous symptoms. In children, the term CNO, or more specifically,
chronic nonbacterial osteomyelitis, supersedes the former name CRMO (chronic recurrent
multifocal osteomyelitis), acknowledging that not all patients meet the criteria for
multifocality or recurrence. Although the etiology of CNO remains elusive, it is perceived
as an autoinflammatory disease stemming from dysregulation of the innate immune system.[56 ]
[57 ]
Pediatric CNO, affecting children and adolescents, commonly affects the metaphyses
of lower limb long bones (including those of the knee joint). In adults, where metaphyses
are closed, the disease predominantly manifests in the anterior chest wall (specifically
the sternoclavicular and sternomanubrial joints), the spine, and sacroiliac joints;
peripheral joint involvement is rare.
Both pediatric and adult forms are characterized by excessive osteitis and periostitis,
sometimes leading to peculiar new bone formation. Radiographs initially display bone
lucency, progressing to sclerosis and new bone formation. Although MRI reveals active
inflammation, it often fails to show periosteal new bone formation. Nevertheless,
MRI, particularly WB-MRI, is the preferred method for early diagnosis in both children
and adults ([Fig. 12 ]).
Fig. 12 A young boy with knee pain and chronic nonbacterial osteitis. (a, b ) Radiography shows a lytic lesion associated with the epiphysis in the distal femur
(white arrowheads) that is also visible in (c ) magnetic resonance imaging (MRI) T1 and shows (d ) surrounding bone marrow edema in the proton-density fat-suppressed sequence (open
arrows). (e, f ) Follow-up whole-body MRI reveals further lesions of the contralateral knee, the
greater trochanter, and the acetabulum (black arrowheads) and another lesion of the
thoracic spine with vertebral collapse (black arrow).
In adult CNO, FDG-PET/CT has been used to evaluate sterile osteitis, and it can assist
in evaluating treatment response by identifying changes in FDG uptake and comparing
SUVs. Although frequently recommended and occasionally indispensable, bone biopsy
to rule out septic inflammation and abscesses is not always necessary.
Conclusion
Over the past few decades, interest has been growing in novel imaging techniques for
the assessment of inflammatory arthropathies.
In the context of suspected gouty arthritis, both US and DECT enable early diagnosis,
each without proven superiority for initial assessment. Cost-effective US can reveal
crystal depositions undetectable by DECT (e.g., the double-contour sign) and active
inflammation manifested as synovitis with vascularization currently evaluated with
even more sensitivity, thanks to microflow vascularity options. US also facilitates
the planning or guidance of fluid aspiration or joint infiltration. Conversely, DECT
offers standardized assessments suitable for follow-up, unveils tophi in areas inaccessible
to US, and detects BME, potentially indicating active inflammation. DECT is also used
for monitoring patients and evaluating treatment response. Both modalities are sensitive
to bone erosion, a relatively late occurring but significant indicator of inflammatory
arthritides.
In patients with suspected RA of the knee, US and MRI offer a sensitive assessment
of active and subclinical inflammation. MRI is easier to standardize and used increasingly
in clinical trials, but more feasible US scores are in development to compensate this
shortfall.[58 ]
[59 ]
For all autoimmune and/or autoinflammatory diseases, new imaging techniques are emerging,
providing both qualitative and quantitative assessment of bone marrow, cartilage,
tendons, and superficial tissues. As we have outlined in this review, many of these
methods have been used successfully in the research setting. Others have already been
incorporated in diagnostic algorithms, whereas incorporation of the remaining in clinical
practice has been limited, and many challenges remain to be addressed.
