Annals of Physical and Rehabilitation Medicine 59 (2016) 161–169

Available online at

ScienceDirect
www.sciencedirect.com

Review

Imaging for osteoarthritis

D. Hayashi a, F.W. Roemer a,b, A. Guermazi a,*
a
Department of Radiology, Boston University School of Medicine, 820 Harrison Avenue, FGH Building 3rd Floor, Boston, MA 02118, United States
b
Department of Radiology, University of Erlangen-Nuremberg, D-91012 Erlangen, Germany

A R T I C L E I N F O A B S T R A C T

Article history: Osteoarthritis (OA) is a widely prevalent disease worldwide and, with an increasing ageing society, is a
Received 1st November 2015 challenge for the field of physical and rehabilitation medicine. Technologic advances and implementa-
Accepted 17 December 2015 tion of sophisticated post-processing instruments and analytic strategies have resulted in imaging
playing a more and more important role in understanding the disease process of OA. Radiography is still
Keywords: the most commonly used imaging modality for establishing an imaging-based diagnosis of OA. The need
Osteoarthritis for an effective non-surgical OA treatment is highly desired, but despite on-going research efforts no
Imaging
disease-modifying OA drugs have been discovered or approved to date. MR imaging-based studies have
Radiography
MR imaging
revealed some of the limitations of radiography. The ability of MR to image all relevant joint tissues
Ultrasound within the knee and to visualize cartilage morphology and composition has resulted in MRI playing a key
CT role in understanding the natural history of the disease and in the search for new therapies. Our review
PET will focus on the roles and limitations of radiography and MRI with particular attention to knee OA. The
use of other modalities (e.g. ultrasound, nuclear medicine, computed tomography (CT), and CT/MR
arthrography) in clinical practice and OA research will also be briefly described. Ultrasound may be
useful to evaluate synovial pathology in osteoarthritis, particularly in the hand.
ß 2015 Elsevier Masson SAS. All rights reserved.

Knee osteoarthritis is a major public health problem that
primarily affects the elderly. Almost 10% of the United States
population suffers from symptomatic knee osteoarthritis by the
age of 60 [S1]. In total, the health care expenditures of this
condition have been estimated at US$ 186 billion annually
[S2]. Despite this there are no approved interventions that
ameliorate structural progression of this disorder.
The increasing importance of imaging in osteoarthritis for
diagnosis, prognostication and follow-up is well recognized by
both clinicians and osteoarthritis researchers. While conventional
radiography is the gold standard imaging technique for the
evaluation of known or suspected osteoarthritis in clinical practice
and research, it has limitations that have become apparent in the
course of large magnetic resonance imaging (MRI)-based knee
osteoarthritis studies [1,2]. Pathological changes may be evident in
all structures of a joint with osteoarthritis although traditionally
researchers have viewed articular cartilage as the central feature
and as the primary target for intervention and measurement. Of
the commonly employed imaging techniques, only MRI can assess
all structures of the joint, including cartilage, meniscus, ligaments,
* Corresponding author. Tel.: +16174143893.
E-mail address: guermazi@bu.edu (A. Guermazi).
muscle, subarticular bone marrow and synovium, and thus can
show the knee as a whole organ three-dimensionally. In addition, it
can directly help in the assessment of cartilage morphology and
composition. The advantages and limitations of conventional
radiography, MRI and other techniques such as ultrasound, nuclear
medicine, computed tomography (CT) and CT arthrography in the
imaging of osteoarthritis in both clinical practice and research are
described in this review article.
1. Conventional radiography
Radiography is the simplest, least expensive and most
commonly deployed imaging modality for OA. It enables detection
of OA-associated bony features such as osteophytes, subchondral
sclerosis and cysts [3]. Radiography can also determine joint space
width (JSW), which is a surrogate for cartilage thickness and
meniscal integrity in knees, but direct visualization of these
articular structures is not possible. Despite this limitation, slowing
of radiographically detected joint space narrowing (JSN) remains
the only structural end point currently approved by the U.S. Food
and Drug Administration to demonstrate efficacy of disease-
modifying OA drugs in phase III clinical trials. OA is radiographi-
cally defined by the presence of marginal osteophytes

http://dx.doi.org/10.1016/j.rehab.2015.12.003
1877-0657/ß 2015 Elsevier Masson SAS. All rights reserved.
162 D. Hayashi et al. / Annals of Physical and Rehabilitation Medicine 59 (2016) 161–169

Fig. 1. The Kellgren-Lawrence classification is a composite scale of OA severity taking into account primarily the radiographic OA features of marginal osteophytes and joint
space narrowing in the AP radiograph. A. Kellgren-Lawrence grade 1. Minimal, equivocal osteophytes are observed at the medial joint margins (large arrows). Note that, so-
called notch osteophytes at the centre of the joint (small arrow) are not considered in the Kellgren-Lawrence scale. B. Kellgren-Lawrence grade 2 is characterized by presence
of at least one definite marginal osteophyte (arrow) without evidence of joint space narrowing. C. Kellgren-Lawrence grade 3 knees exhibit signs of definite joint space
narrowing (black arrows) and marginal osteophytes (white arrows). The amount of joint space narrowing is not taken into account. D. Kellgren-Lawrence grade 4 is defined by
bone-to-bone contact and complete obliteration of the joint space (black arrows). Note definite marginal osteophytes in addition (white arrows).

[S3]. Worsening of JSN is the most commonly used criterion for the tibiofemoral, lateral tibiofemoral, and patellofemoral). A recent
assessment of structural OA progression and the total loss of JSW study using data from the OA Initiative and the OARSI atlas for
(‘‘bone-on-bone’’ appearance) is one of the indicators for joint semiquantitative grading of JSN demonstrated that centralized
replacement. radiographic reading is important from the point of observer
We now know cartilage loss is not the only contributor to JSN reliability, as even expert readers seem to apply different
but that changes in the meniscus such as meniscal extrusion and thresholds for JSN grading [7].
meniscal substance loss are also causative factors. The lack of
sensitivity and specificity of radiography for the detection of OA- 1.2. Quantitative assessments
associated articular tissue damage, and its poor sensitivity to
change longitudinally are other limitations of radiography. JSW is the distance between the projected femoral and tibial
Changes in knee positioning can also be problematic in longitu- margins on the anteroposterior radiographic image. Measure-
dinal studies and can affect the quantitative measurement of ments may be performed manually or in a semi-automated fashion
various radiographic parameters including JSW. Despite these using computer software. Quantification using image processing
limitations, radiography is still the gold standard for establishing software requires a digital image, whether digitized plain films or
an imaging-based diagnosis of OA and for assessment of structural images acquired using fully digital modalities such as computed
modification in clinical trials of knee OA. radiography and digital radiography. Minimum JSW is the
standard metric, but the use of location-specific JSW has also
1.1. Semiquantitative assessments been reported [S4]. Using software analysis of digital knee
radiographic images, measures of location-specific JSW were
The severity of radiographic OA can be assessed with shown to be comparable with MR imaging in detecting OA
semiquantitative scoring systems. The Kellgren and Lawrence progression [8]. Various degrees of responsiveness have been
(KL) grading system [4] is a widely accepted scheme for defining
radiographic OA based on the presence of a definite osteophyte
(grade 2). However, KL grading has its limitations; in particular, KL
grade 3 includes all degrees of JSN, regardless of the actual extent.
Fig. 1 depicts representative examples of the different KL grades as
shown on the AP radiograph. Recently, the so-called atrophic
phenotype of knee OA that is characterized by definite joint space
narrowing without concomitant osteophyte formation has gained
increasing attention as a potential risk factor for more rapid
progressive OA, which may be considered a potential adverse event
in so-called anti-nerve growth factor (NGF) drug trials, a class of
new promising antianalgesic compounds currently under investi-
gation [5]. Although this phenotype is rare, it needs special
attention in the research community as it potentially also is a
reflection of more rapid disease progression [6]. Fig. 2 shows an
example of the atrophic phenotype of radiographic OA.
The Osteoarthritis Research Society International (OARSI) atlas
[3] provides image examples for grades for specific features of OA
Fig. 2. Kellgren-Lawrence 3 knees with definite joint space narrowing (arrowheads)
rather than assigning global scores according to definitions like the but only minimal osteophyte formation (arrow) are considered to represent the so-
KL grading system. The atlas grades tibiofemoral JSW and called atrophic phenotype of knee OA. This subtype of radiographic knee OA may be
osteophytes separately for each compartment of the knee (medial a reflection of more rapid disease progression.
 D. Hayashi et al. / Annals of Physical and Rehabilitation Medicine 59 (2016) 161–169
observed depending on the degree of OA severity, length of the
follow-up period, and the knee positioning protocol [S4] [8]. Mea-
surements of JSW obtained from knee radiographs have been
found to be reliable, especially when the study lasted longer than
two years and when the radiographs were obtained with the knee
in a standardized flexed position [S5].
The role of malalignment in OA disease process has been
described in the literature. A clinical trial showed that varus
malalignment negated the slowing of structural progression of
medial JSN by doxycycline [S6]. Valgus malalignment of the lower
limb was shown to increase the risk of disease progression in knees
with radiographic lateral knee OA [9]. Malalignment might be a
target for prevention of radiographic knee OA as determined by JSN
in overweight and obese women [10].
2. Other radiography-based techniques
Interestingly, two older methods – tomosynthesis and bone
texture analysis – have experienced a revival recently. Tomosyn-
thesis generates an arbitrary number of cross-sectional images
from a single pass of the X-ray tube. A recent study showed that
tomosynthesis is more sensitive in the detection of osteophytes
and subchondral cysts than radiography, using 3T MRI as the
reference [11]. Moreover, tomosynthesis was shown to offer
excellent intrareader reliability regardless of the reader experience
[12]. Quantification of JSW using tomosynthesis has also been
reported [S7]. Bone texture analysis extracts from conventional
radiographs information on two-dimensional trabecular bone
texture that relates directly to three-dimensional bone structure. It
has been shown that bone texture may be a predictor of
progression of tibiofemoral OA [13].
3. MRI
Because of the high cost, MRI is not routinely used in clinical
management of OA patients. However, MRI has become a key-
imaging tool for OA research [14–16] thanks to its ability to assess
pathology in structures not visualized by radiography i.e. articular
cartilage, menisci, ligaments, synovium, capsular structures, fluid
collections and bone marrow [2,17–21]. With MRI the joint can be
evaluated as a whole organ and multiple tissue changes can be
monitored simultaneously; pathologic changes of pre-radiograph-
ic OA can be detected; and biochemical changes within joint
tissues such as cartilage can be assessed before morphologic
changes become evident. An MRI-based definition of OA is
available [22]. In addition, with MRI, OA can be classified into
hypertrophic and atrophic phenotypes, according to the size of
osteophytes and concomitant presence or absence of JSN
[6]. Importantly, the use of MRI has led to significant findings
about the association of pain with bone marrow lesions [23] and
synovitis [24], with implications for future OA clinical trials. Fig. 3
shows an example of typical large bone marrow lesions that are
commonly seen in conjunction with severe cartilage damage and
may be responsible for pain in OA.
3.1. Technical considerations
It should be emphasized that investigators need to select
appropriate MR pulse sequences for the purpose of each study. For
example, focal cartilage defects and bone marrow lesions are best
assessed using fluid-sensitive fast spin echo sequences (e.g. T2-
weighted, proton density-weighted or intermediate-weighted)
with fat suppression [14,17]. Moreover, MRI may sometimes be
affected by artifacts that mimic pathological findings. For example,
susceptibility artifacts can be misinterpreted as cartilage loss or
meniscal tear. Gradient recalled-echo sequences are known to be
163
Fig. 3. Subchondral bone marrow lesions (BMLs) are commonly observed in knees
with OA. These are a reflection of increased loading and particularly large BMLs may
be responsible for pain and structural progression. BMLs are visualized as ill-
defined areas of hyperintensity in the subchondral bone (arrows).
particularly prone to this type of artifact [25]. To ascertain optimal
assessment of MRI-derived data, trained musculoskeletal radio-
logists should be consulted when designing imaging-based OA
studies or when interpreting data from the studies.
3.2. Semiquantitative MRI scoring systems for knee OA
A detailed review of semiquantitative MRI assessment of OA has
been published [15]. In addition to the three well-established
scoring systems – the Whole Organ Magnetic Resonance Imaging
Score (WORMS) [26], the Knee Osteoarthritis Scoring System
(KOSS) [S8], and the Boston Leeds Osteoarthritis Knee Score
(BLOKS) [S9] – a new scoring system called the MR Imaging
Osteoarthritis Knee Score (MOAKS) [27] was most recently
published. Of the three systems, WORMS and BLOKS have been
widely disseminated and used, but these scoring systems both
have strengths and weaknesses. MOAKS provides a refined scoring
tool for cross-sectional and longitudinal semiquantitative MRI
assessment of knee OA. It includes semiquantitative scoring of
pathological features: bone marrow lesions; subchondral cysts;
articular cartilage; osteophytes; Hoffa synovitis and synovitis-
effusion; meniscus; tendons and ligaments; and periarticular
features such as cysts and bursitides.
The use of within-grade changes for longitudinal assessment is
commonly applied in OA research using semiquantitative MRI
approaches [28]. Within-grade scoring describes progression or
improvement of a lesion that does not meet the criteria of a full
grade change but does represent a definite visual change in
comparison to the previous visit. A recent study demonstrated that
within-grade changes in semiquantitative MRI assessment of
cartilage and bone marrow lesions are valid and their use may
increase the sensitivity of semiquantitative readings for detecting
longitudinal changes in these structures [28].
Synovitis is an important feature of OA and is associated with
pain [24]. Although synovitis can be evaluated with non-contrast-
enhanced MR imaging by using the presence of signal changes in
the Hoffa fat pad or joint effusion as an indirect marker of synovitis,
164 D. Hayashi et al. / Annals of Physical and Rehabilitation Medicine 59 (2016) 161–169

Fig. 4. Comparison of inflammatory manifestations of disease using non-enhanced and contrast-enhanced sequences. A. Axial intermediate-weighted fat-suppressed MRI
shows homogeneous hyperintensity within the joint cavity consistent with grade 2 effusion-synovitis by MOAKS and WORMS (asterisk). Note: BML in the posterior lateral
femoral condyle consistent with traction edema at the insertion of the anterior cruciate ligament (arrows). B. Corresponding T1-weighted fat-suppressed image after
intravenous contrast administration clearly differentiates between intra-articular joint fluid depicted as hypointensity (asterisk) and true synovial thickening visualized as
hyperintense contrast enhancement of the synovial membrane (arrows). Note that BMLs are similarly depicted on T2-weighted fat-suppressed and T1-weighted contrast-
enhanced fat-suppressed MRI (arrowheads).

only contrast-enhanced MR imaging can reveal the true extent of
synovial inflammation [29]. Scoring systems of synovitis based on
contrast-enhanced MR imaging have been published [24], and
these could potentially be used in clinical trials of new OA drugs
that target synovitis. Fig. 4 shows a direct comparison between
non-enhanced and contrast-enhanced MRI for the visualization of
synovitis and joint effusion.
Other available semiquantitative MRI scoring systems related
to knee OA include Cartilage Repair Osteoarthritis Knee Score
(CROAKS) [30] and Anterior Cruciate Ligament Osteoarthritis Score
(ACLOAS) [31], which can be applied to research studies focusing
on these structures. Specifically for imaging of cartilage repair
tissue, MRI Observation Cartilage Repair Tissue (MOCART) scoring
is available and has been applied to research studies involving
cartilage repair surgery [32].
3.3. Semiquantitative MRI scoring system for hand OA
Radiography is still the modality of choice for assessment of
hand OA in a clinical setting but the use of more sensitive imaging
techniques such as ultrasound and MRI is becoming more common
in OA research [S10]. A semiquantitative MRI scoring system for
hand OA features called the OMERACT Hand Osteoarthritis
Magnetic Resonance Scoring System (HOAMRIS) is available and
offers high reliability and responsiveness [S11]. Scored pathologi-
cal features include synovitis, erosions, cysts, osteophytes, JSN,
malalignment and bone marrow lesions. Using these scoring
systems, Haugen et al. showed that MRI could detect approxi-
mately twice as many joints with erosions and osteophytes as
conventional radiography (P < 0.001) [33] and that moderate/
severe synovitis, bone marrow lesions, erosions, attrition and
osteophytes were associated with joint tenderness independently
of each other [34]. These studies demonstrated that some of the
semiquantitatively assessed MRI features of hand OA may be
potential targets for therapeutic interventions.
Scoring System (HOAMS) is available for use in observational
studies and clinical trials of hip joints [35]. In HOAMS, fourteen
articular features are assessed: cartilage morphology, subchondral
bone marrow lesions, subchondral cysts, osteophytes, acetabular
labrum, synovitis (scored only when contrast-enhanced sequences
are available), joint effusion, loose bodies, attrition, dysplasia,
trochanteric bursitis/insertional tendonitis of the greater trochan-
ter, labral hypertrophy, paralabral cysts and herniation pits at the
supero-lateral femoral neck. HOAMS demonstrated satisfactory
reliability and good agreement concerning intra- and inter-
observer assessment. Recently another system was introduced
that uses a similar approach [36]. Direct comparisons of HOAMS
and SHOMRI in regard to responsiveness and validity are missing
to date.
3.5. Quantitative analysis of articular cartilage and other tissues
Quantitative measurement of cartilage morphology segments
the cartilage image and exploits the three-dimensional nature of
MRI data sets to evaluate tissue dimensions (such as thickness and
volume) or signal as continuous variables. Examples of nomencla-
ture for MRI-based cartilage measures and strategies for efficient
analysis of longitudinal changes in subregional cartilage thickness
were proposed by Eckstein et al. [37] [S12]. Quantitative cartilage
morphometry has been widely applied in OA studies [S13]
[38]. Quantitative measures of articular cartilage structure, such
as cartilage thickness loss and denuded areas of subchondral bone
have been shown to predict an important clinical outcome, i.e.
knee replacement [S13]. Quantitative MRI analysis can also be
applied to evaluate non-cartilage tissues in the joint including
menisci [S14], bone marrow lesions [S15], synovitis [S16] and joint
effusion [S17]. However, using segmentation approaches for ill-
defined lesions such as bone marrow lesions is more challenging
than segmentation of clearly delineated structures such as
cartilage, menisci and effusion.

3.6. Compositional MRI

3.4. Semiquantitative MRI scoring system for hip OA
The hip joint has a spherical structure and its very thin covering
of articular hyaline cartilage makes MRI assessment of the hip
much more difficult than the knee. A whole organ semiquantitative
multi-feature scoring system called the Hip Osteoarthritis MRI
Compositional MRI allows visualization of the biochemical
properties of different joint tissues. It may be sensitive to early,
pre-morphologic changes that cannot be visualized on conven-
tional MRI. Compositional MRI has been mostly applied for
ultrastructural assessment of catilage, although the technique can
 D. Hayashi et al. / Annals of Physical and Rehabilitation Medicine 59 (2016) 161–169 165

also be used to assess other tissues such as the menisci or
ligaments. Detailed reviews on this topic have been published
recently [39,40]. Compositional MRI of cartilage matrix changes
can be performed using advanced MRI techniques such as delayed
gadolinium-enhanced MRI of cartilage (dGEMRIC), T1 rho, and T2
mapping [S18]. Of these, the first two take advantage of the
concentration of highly negatively charged glycosaminoglycans
(GAGs) in healthy hyaline cartilage; loss of these GAGs in focal
areas affected by possible early disease can be visualized. Both
dGEMRIC and T1 rho focus on charge density in cartilage. In
contrast, T2 concentrations are affected by a complex combination
of collagen orientation and hydration of cartilage.
Compositional MRI techniques are currently not routinely used
in clinical practice and remain research tools that are available only
at a limited number of institutions. Nevertheless, they have been
applied in clinical trials and observational studies. A recent
placebo-controlled double-blind pilot study of collagen hydroly-
sate for mild knee OA demonstrated that the dGEMRIC score
increased (meaning higher GAG content and better cartilage
status) in tibial cartilage regions of interest in patients receiving
collagen hydrolysate, and decreased in the placebo group, with a
significant difference being observed at 24 weeks [41]. Crema et al.
showed that a decrease in dGEMRIC indices was associated with an
increase in cartilage thickness in the medial compartment of the
knee, suggesting that an increase in cartilage thickness may also be
related to a decrease in proteoglycan concentration [42]. Souza
et al. [43] demonstrated that acute loading of the knee joint
resulted in a significant decrease in T1 rho and T2 relaxation times
of the medial tibiofemoral compartment and especially in cartilage
regions with small focal defects, suggesting that changes of T1 rho
values under mechanical loading may be related to the
biomechanical and structural properties of cartilage. Hovis et al.
reported that light exercise was associated with low cartilage
T2 values but moderate and strenuous exercise was associated
with high T2 values in women, suggesting that activity levels can
affect cartilage composition [44]. In an interventional study
assessing the effect of weight loss on articular cartilage,
Anandacoomarasamy et al. reported that improved articular
cartilage quality was reflected as an increase in the dGEMRIC
index over one year for the medial but not the lateral compartment
[45], highlighting the role of weight loss in possible clinical and
structural improvement. Fig. 5 shows an example of a knee with a
prevalent meniscal tear and corresponding decrease in GAG
content in the central aspect of the cartilage of the central medial
femoral condyle.
Novel compositional techniques have been further explored,
including in vivo diffusion tensor imaging [S19], T2* mapping [46],
ultra-short echo time-enhanced (UTE) T2* mapping [47], and
sodium imaging [48]. These techniques show promise, but are still
at an early stage of research and development.
4. Ultrasound
Ultrasound imaging enables real time, multiplanar imaging at
relatively low cost. It offers reliable assessment of OA-associated
features, including inflammatory and structural abnormalities,
without contrast administration or exposure to radiation [49]. Limi-
tations of ultrasound include that it is an operator-dependent
technique and that the physical properties of sound limit its ability
to assess deep articular structures and the subchondral bone.
Ultrasound is useful for evaluation of cortical erosive changes
and synovitis in inflammatory arthritis. In OA, the major advantage
of ultrasound over conventional radiography is the ability to detect
synovial pathology. Current generation ultrasound technology can
detect synovial hypertrophy, increased vascularity and the
presence of synovial fluid in joints affected by OA [49]. The
Outcome Measures in Rheumatoid Arthritis Clinical Trials (OME-
RACT) Ultrasonography Taskforce defined synovial hypertrophy on
ultrasound as ‘‘abnormal hypoechoic (relative to subdermal fat,
but sometimes may be isoechoic or hyperechoic) intra-articular
tissue that is non-displaceable and poorly compressible and which
may exhibit Doppler’’ [S20]. Although this definition was devel-
oped for use in rheumatoid arthritis, it may also be applied to OA
because the difference in synovial inflammation between OA and
rheumatoid arthritis is largely quantitative rather than qualitative
[49].
A preliminary ultrasonographic scoring system for features of
hand OA published in 2008 [50] included evaluation of grey-scale
synovitis and power Doppler signal in 15 joints of the hand. These
features were assessed for their presence/absence and if present
were scored semiquantitatively using a 1–3 scale. Overall, the
reliability exercise demonstrated moderately good intra- and
inter-reader reliability. This study showed that an ultrasound
outcome measure suitable for multicenter trials assessing hand OA
is feasible and likely to be reliable, and has provided a foundation
for further development.

Fig. 5. Compositional MRI. A. Sagittal fat-suppressed proton density-weighted image shows a horizontal meniscal tear at the posterior horn of the medial meniscus reaching
the undersurface of the mensicus (arrow). B. Corresponding color-coded T1-weighted dGEMRIC image after intravenous contrast administration shows a decrease of the
dGEMRIC index in the weight bearing part of the medial femoral condyle coded in red (arrows) representing early cartilage degeneration reflected as a decrease in
glucosaminoglycan content compared to the posterior parts of the femoral cartilage that are coded in yellow and green.
166 D. Hayashi et al. / Annals of Physical and Rehabilitation Medicine 59 (2016) 161–169
Fig. 6. Assessment of meniscal extrusion by ultrasound. A. Coronal ultrasound screenshot of the medial tibiofemoral compartment shows an extruded body of the medial
meniscus under weight bearing conditions (arrows). B Corresponding coronal fat-suppressed proton density-weighted MRI shows extrusion to a lesser extent (arrows) due to
the prone positioning for the MRI acquisition.
Ultrasound has been increasingly deployed for assessment of
hand OA. Kortekaas et al. showed that ultrasound-detected
osteophytes and JSN are associated with hand pain [51]. The same
group of investigators also showed that ultrasound-detected signs
of inflammation are associated with development of erosions in
hand OA, implicating inflammation plays a role in its pathogenesis
and could be a therapeutic target [S21]. Ultrasound can be used to
evaluate the therapeutic efficacy. A decrease in pain correlated
with a decrease in synovial thickening and power Doppler
ultrasound score before and after weekly ultrasound-guided
intra-articular injections of hyaluronic acid [52]. Real time fusion
of ultrasound and MR imaging in hand and wrist OA have been
attempted, and found a high concordance of the bony profile
visualization at the level of osteophytes [S10].
Ultrasound has also been used to evaluate features of knee OA
and hip OA. A cross-sectional, multicenter European study
analyzed 600 patients with painful knee OA, and found ultra-
sound-detected synovitis correlated with advanced radiographic
OA and clinical symptoms and signs suggestive of an inflammatory
‘‘flare’’ [53]. Ultrasound-detected inflammatory features were
positively and linearly associated with knee pain in motion in
patients with equal radiographic grades of knee OA in both knees
[54], supporting the association between synovitis and knee pain,
which has also been reported in MR imaging-based studies [24].
Fig. 6 shows an example of the utility of ultrasound for the
visualization of meniscal extrusion. An advantage of ultrasound
over MRI in regard to extrusion measurements is the possibility of
weight bearing examinations [55].
5. Nuclear medicine
Nuclear medicine imaging with radiotracers enables imaging of
active metabolism and visualization of bone turnover changes seen
with osteophyte formation, subchondral sclerosis, subchondral
cyst formation and bone marrow lesions as well as sites of
synovitis [S22]. Scintigraphy with 99mTc-hydroxymethane diphos-
phonate (HDP) and positron emission tomography (PET) with
2-18F-fluoro-2-deoxy–D-glucose (18FDG) or 18F-fluoride (18F )
have been used to assess OA. Bone scintigraphy can provide a
full-body survey that helps to discriminate between soft tissues
and bone origin of pain, and to locate the site of pain in patients
with complex symptoms [S23]. 18FDG-PET can demonstrate the
site of synovitis and bone marrow lesions associated with OA
[56]. Additionally, there have been increasing research efforts to
apply SPECT/CT for imaging of osteoarthritis [57]. A major
limitation of radioisotope methods is poor anatomical resolution
but there are ways to overcome it. Hybrid technologies such as
PET-CT and PET-MR imaging combine functional imaging with
high-resolution anatomical imaging. Fig. 7 shows an example of
PET-CT in knee OA that combines the high resolution of the CT with
the metabolic information of the PET. However, nuclear medicine
imaging is not commonly applied to imaging of OA in a routine
clinical setting.
6. CT
CT is excellent for depicting cortical bone and soft tissue
calcifications, and has an established role in assessing facet joint
OA of the spine. Using a CT-based semiquantitative grading system
of facet joint OA, studies have shown a high prevalence of facet
joint OA which increases with age, and is most common at the L4–
L5 spinal level [58] as well as associations between obesity with
higher prevalence of facet joint OA and that between increasing
age with higher prevalence of disc narrowing, facet joint OA, and
degenerative spondylolisthesis [S24]. Ionizing radiation and
limited ability of soft tissue evaluation limit the routine use of
CT in clinical imaging of OA.
7. CT and MR arthrography
With CT or MR arthrography, assessment of cartilage damage is
possible with high anatomical resolution in a multiplanar fashion
[S23] [59,60]. CT arthrography is currently the most accurate
method for evaluating superficial and focal cartilage damage,
offering high spatial resolution and high contrast between the low-
attenuating cartilage and high-attenuating superficial (contrast
material filling the joint space) and deep (subchondral bone)
boundaries [S23]. Fig. 8 shows an example of CT arthrography that
delineates superficial and full thickness cartilage damage in
excellent fashion. CT can also depict subchondral bone sclerosis
and osteophytes. For subchondral changes, MR arthrography
allows delineation of subchondral bone marrow lesions on the
fluid-sensitive sequences with fat suppression [S23]. Both techni-
ques enable visualization of central osteophytes, which are
 D. Hayashi et al. / Annals of Physical and Rehabilitation Medicine 59 (2016) 161–169 167

Fig. 7. PET-CT. A. Conventional coronal CT reformation shows definite OA on the left with marginal tibial and femoral osteophytes (thin arrows) and definite medial
tibiofemoral joint space narrowing. Total knee arthroplasty of the right knee with the metal implants clearly depicted as markedly hyperdense structures (large arrows). Note
beam-hardening artifacts due to the metal hardware particularly surrounding the distal femur. B. Color-coded fusion image of PET and CT shows marked tracer uptake in the
perimensical region bilaterally representing active synovitis (large arrows). C. Corresponding axial fusion image shows synovitis also in the peripatellar recesses (small
arrows). In addition, there is marked synovitis posteriorly adjacent to the posterior cruciate ligament, the most common location of synovitis in knee OA (large arrow). To date
PET-CT does not play a relevant role for the assessment of knee OA structural changes in the clinical routine practice.

8. Conclusion

Imaging of OA has helped investigators to understand risk factors

Fig. 8. CT arthrography is considered the imaging gold standard for the visualization
of the articular surface integrity. After intra-articular injection of an iodine-based
contrast agent the cartilage surface can be depicted as a hypodense structure. Image
example shows a focal full thickness defect at the lateral tibia (arrow).
for disease onset and progression and the interactions between
different joint tissues and peritarticular tissue changes. Despite
those research advances the role of imaging in the routine clinical
management of OA patients remains less well-defined in that
imaging findings do not always correlate with clinical symptoms.
Given such limitations, radiography is still most commonly used for
semiquantitative and quantitative evaluation of structural OA
features. MRI is currently the most important imaging modality
for OA research, and available options include semiquantitative,
quantitative and compositional analyses. Ultrasound is commonly
used in hand OA studies and is particularly useful for evaluation of
synovitis and for guidance of joint-related procedures. Nuclear
medicine, CT and CT-MR arthrography have limited roles compared
to radiography, MRI and ultrasound.
Role of the funding source

No funding received.

associated with more severe changes of OA than marginal Disclosure of interests

osteophytes [S25]. Due to the high cost, invasive nature and
potential risk, albeit low, associated with intra-articular injection, Dr. Guermazi has received consultancies, speaking fees, and/or
arthrographic examinations are rarely used in large scale clinical or honoraria from TissueGene, OrthoTrophix, and Merck Serono and
epidemiological OA studies. is the President of Boston Imaging Core Lab (BICL), a company
168 D. Hayashi et al. / Annals of Physical and Rehabilitation Medicine 59 (2016) 161–169
providing image assessment services. Dr. Roemer is Chief Medical
Officer and shareholder of BICL. Dr. Hayashi declares that he has no
competing interest.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.rehab.2015.12.
003.
References
[1] Guermazi A, Hayashi D, Roemer F, Felson DT, Wang K, Lynch J, et al. Severe
radiographic knee osteoarthritis – does Kellgren and Lawrence grade 4 repre-
sent end stage disease? Osteoarthritis Cartilage 2015;23:1499–505.
[2] Guermazi A, Niu J, Hayashi D, Roemer FW, Englund M, Neogi T, et al. Prevalence
of abnormalities in knees detected by MRI in adults without knee osteoarthri-
tis: population based observational study (Framingham Osteoarthritis Study).
BMJ 2012;345:e5339.
[3] Altman RD, Gold GE. Atlas of individual radiographic features in osteoarthritis,
revised. Osteoarthritis Cartilage 2007;15(Suppl. A):A1–56.
[4] Kellgren JH, Lawrence JS. Radiological assessment of osteo-arthrosis. Ann
Rheum Dis 1957;16:494–502.
[5] Roemer FW, Hayes CW, Miller CG, Hoover K, Guermazi A. Imaging atlas for
eligibility and on-study safety of potential knee adverse events in anti-NGF
studies (Part 1). Osteoarthritis Cartilage 2015;23(Suppl. 1):S22–42.
[6] Roemer FW, Guermazi A, Niu J, Zhang Y, Mohr A, Felson DT. Prevalence of
magnetic resonance imaging-defined atrophic and hypertrophic phenotypes
of knee osteoarthritis in a population-based cohort. Arthritis Rheum
2012;64:429–37.
[7] Guermazi A, Hunter DJ, Li L, Benichou O, Eckstein F, Kwoh CK, et al. Different
thresholds for detecting osteophytes and joint space narrowing exist between
the site investigators and the centralized reader in a multicenter knee osteo-
arthritis study – data from the Osteoarthritis Initiative. Skeletal Radiol
2012;41:179–86.
[8] Duryea J, Neumann G, Niu J, Totterman S, Tamez J, Dabrowski C, et al.
Comparison of radiographic joint space width with magnetic resonance
imaging cartilage morphometry: analysis of longitudinal data from the
Osteoarthritis Initiative. Arthritis Care Res (Hoboken) 2010;62:932–7.
[9] Felson DT, Niu J, Gross KD, Englund M, Sharma L, Cooke TD, et al. Valgus
malalignment is a risk factor for lateral knee osteoarthritis incidence and
progression: findings from MOST and the Osteoarthritis Initiative. Arthritis
Rheum 2013;65:355–62.
[10] Runhaar J, van Middelkoop M, Reijman M, Vroegindeweij D, Oei DH, Bierma-
Zeinstra SM. Malalignment: a possible target for prevention of incident knee
osteoarthritis in overweight and obese women. Rheumatology (Oxford)
2014;53:1618–24.
[11] Hayashi D, Xu L, Roemer FW, Hunter DJ, Li L, Katur AM, et al. Detection of
osteophytes and subchondral cysts in the knee with use of tomosynthesis.
Radiology 2012;263:206–15.
[12] Hayashi D, Xu L, Gusenburg J, Roemer FW, Hunter DJ, Li L, et al. Reliability of
semiquantitative assessment of osteophytes and subchondral cysts on tomo-
synthesis images by radiologists with different levels of expertise. Diagn Interv
Radiol 2014;20:353–9.
[13] Woloszynski T, Podsiadlo P, Stachowiak GW, Kurzynski M, Lohmander LS,
Englund M, et al. Prediction of progression of radiographic knee osteoarthritis
using tibial trabecular bone texture. Arthritis Rheum 2012;64:688–95.
[14] Hayashi D, Guermazi A, Roemer FW. MRI of osteoarthritis: the challenges of
definition and quantification. Semin Musculoskelet Radiol 2012;16:419–30.
[15] Guermazi A, Roemer FW, Haugen IK, Crema MD, Hayashi D. MRI-based
semiquantitative scoring of joint pathology in osteoarthritis. Nat Rev Rheu-
matol 2013;9:236–51.
[16] Alizai H, Roemer FW, Hayashi D, Crema MD, Felson DT, Guermazi A. An update
on risk factors for cartilage loss in knee osteoarthritis assessed using MRI-
based semiquantitative grading methods. Eur Radiol 2015;25:883–93.
[17] Hayashi D, Guermazi A, Kwoh CK, Hannon MJ, Moore C, Jakicic JM, et al.
Semiquantitative assessment of subchondral bone marrow edema-like lesions
and subchondral cysts of the knee at 3 T MRI: A comparison between inter-
mediate-weighted fat-suppressed spin echo and Dual Echo Steady State
sequences. BMC Musculoskelet Disord 2011;12:198.
[18] Hayashi D, Englund M, Roemer FW, Niu J, Sharma L, Felson DT, et al. Knee
malalignment is associated with an increased risk for incident and enlarging
bone marrow lesions in the more loaded compartments: the MOST study.
Osteoarthritis Cartilage 2012;20:1227–33.
[19] Roemer FW, Felson DT, Wang K, Crema MD, Neogi T, Zhang Y, et al. Co-
localisation of non-cartilaginous articular pathology increases risk of cartilage
loss in the tibiofemoral joint – the MOST study. Ann Rheum Dis 2013;72:
942–8.
[20] Roemer FW, Kwoh CK, Hannon MJ, Green SM, Jakicic JM, Boudreau R, et al.
Risk factors for magnetic resonance imaging-detected patellofemoral and
tibiofemoral cartilage loss during a six-month period: the joints on glucos-
amine study. Arthritis Rheum 2012;64:1888–98.
[21] Roemer FW, Guermazi A, Felson DT, Niu J, Nevitt MC, Crema MD, et al. Presence
of MRI-detected joint effusion and synovitis increases the risk of cartilage loss
in knees without osteoarthritis at 30-month follow-up: the MOST study. Ann
Rheum Dis 2011;70:1804–9.
[22] Hunter DJ, Arden N, Conaghan P, Eckstein F, Gold G, Grainger A, et al. Definition
of osteoarthritis on MRI: results of a Delphi exercise. Osteoarthritis Cartilage
2011;19:963–9.
[23] Zhang Y, Nevitt M, Niu J, Lewis C, Torner J, Guermazi A, et al. Fluctuation of
knee pain and changes in bone marrow lesions, effusions, and synovitis on
magnetic resonance imaging. Arthritis Rheum 2011;63:691–9.
[24] Guermazi A, Roemer FW, Hayashi D, Crema MD, Niu J, Zhang Y, et al. Assess-
ment of synovitis with contrast-enhanced MRI using a whole-joint semiquan-
titative scoring system in people with, or at high risk of, knee osteoarthritis:
the MOST study. Ann Rheum Dis 2011;70:805–11.
[25] Jarraya M, Hayashi D, Guermazi A, Kwoh CK, Hannon MJ, Moore CE, et al.
Susceptibility artifacts detected on 3 T MRI of the knee: frequency, change
over time and associations with radiographic findings: data from the joints on
glucosamine study. Osteoarthritis Cartilage 2014;22:1499–503.
[26] Peterfy CG, Guermazi A, Zaim S, Tirman PF, Miaux Y, White D, et al. Whole-
Organ Magnetic Resonance Imaging Score (WORMS) of the knee in osteoar-
thritis. Osteoarthritis Cartilage 2004;12:177–90.
[27] Hunter DJ, Guermazi A, Lo GH, Grainger AJ, Conaghan PG, Boudreau RM, et al.
Evolution of semi-quantitative whole joint assessment of knee OA: MOAKS
(MRI Osteoarthritis Knee Score). Osteoarthritis Cartilage 2011;19:990–1002.
[28] Roemer FW, Nevitt MC, Felson DT, Niu J, Lynch JA, Crema MD, et al. Predictive
validity of within-grade scoring of longitudinal changes of MRI-based cartilage
morphology and bone marrow lesion assessment in the tibio-femoral joint –
the MOST Study. Osteoarthritis Cartilage 2012;20:1391–8.
[29] Loeuille D, Sauliere N, Champigneulle J, Rat AC, Blum A, Chary-Valckenaere I.
Comparing non-enhanced and enhanced sequences in the assessment of
effusion and synovitis in knee OA: associations with clinical, macroscopic
and microscopic features. Osteoarthritis Cartilage 2011;19:1433–9.
[30] Roemer FW, Guermazi A, Trattnig S, Apprich S, Marlovits S, Niu J, et al. Whole
joint MRI assessment of surgical cartilage repair of the knee: cartilage repair
osteoarthritis knee score (CROAKS). Osteoarthritis Cartilage 2014;22:779–99.
[31] Roemer FW, Frobel R, Lohmaner LS, Niu J, Guermazi A. Anterior Cruciate
Ligament OsteoArthritis Score (ACLOAS): longitudinal MRI-based whole joint
assessment of anterior cruciate ligament injury. Osteoarthritis Cartilage
2014;22:668–82.
[32] Welsch GH, Zak L, Mamisch TC, Paul D, Lauer L, Mauerer A, et al. Advanced
morphological 3D magnetic resonance observation of cartilage repair tissue
(MOCART) scoring using a new isotropic 3D proton-density, turbo spin echo
sequence with variable flip angle distribution (3D-SPACE) compared to an
isotropic 3D steady-state free precession sequence (True-FISP) and standard
2D sequences. J Magn Reson Imaging 2011;33:180–8.
[33] Haugen IK, Boyesen P, Slatkowsky-Christensen B, Sesseng S, Bijsterbosch J, van
der Heijde D, et al. Comparison of features by MRI and radiographs of the
interphalangeal finger joints in patients with hand osteoarthritis. Ann Rheum
Dis 2012;71:345–50.
[34] Haugen IK, Boyesen P, Slatkowsky-Christensen B, Sesseng S, van der Heijde D,
Kvien TK. Associations between MRI-defined synovitis, bone marrow lesions
and structural features and measures of pain and physical function in hand
osteoarthritis. Ann Rheum Dis 2012;71:899–904.
[35] Roemer FW, Hunter DJ, Winterstein A, Li L, Kim YJ, Cibere J, et al. Hip
Osteoarthritis MRI Scoring System (HOAMS): reliability and associations with
radiographic and clinical findings. Osteoarthritis Cartilage 2011;19:946–62.
[36] Lee S, Nardo L, Kumar D, Wyatt CR, Souza RB, Lynch J, et al. Scoring hip
osteoarthritis with MRI (SHOMRI): a whole joint osteoarthritis evaluation
system. J Magn Reson Imaging 2015;41:1549–57.
[37] Eckstein F, Ateshian G, Burgkart R, Burstein D, Cicuttini F, Dardzinski B, et al.
Proposal for a nomenclature for magnetic resonance imaging-based measures of
articular cartilage in osteoarthritis. Osteoarthritis Cartilage 2006;14:974–83.
[38] Eckstein F, Collins JE, Nevitt MC, Lynch JA, Kraus V, Katz JN, et al. Cartilage
thickness change as an imaging biomarker of knee osteoarthritis progression–
data from the fnih OA biomarkers consortium. Arthritis Rheumatol 2015.
http://dx.doi.org/10.1002/art.39324 [Epub ahead of print].
[39] Guermazi A, Roemer FW, Alizai H, Winalski CS, Welsch G, Brittberg M, et al.
State of the art: MR imaging after knee cartilage repair surgery. Radiology
2015;277:23–43.
[40] Guermazi A, Alizai H, Crema MD, Trattnig S, Regatte RR, Roemer FW. Compo-
sitional MRI techniques for evaluation of cartilage degeneration in osteoar-
thritis. Osteoarthritis Cartilage 2015;23:1639–53.
[41] McAlindon TE, Nuite M, Krishnan N, Ruthazer R, Price LL, Burstein D, et al.
Change in knee osteoarthritis cartilage detected by delayed gadolinium-
enhanced magnetic resonance imaging following treatment with collagen
hydrolysate: a pilot randomized controlled trial. Osteoarthritis Cartilage
2011;19:399–405.
[42] Crema MD, Hunter DJ, Burstein D, Roemer FW, Li L, Eckstein F. Association of
changes in delayed gadoliun-enhanced MRI of cartilage (dGEMRIC) with chan-
ges in cartilage thickness in the medial tibiofemoral compartment of the knee: a
2 year follow-up study using 3.0T MRI. Ann Rheum Dis 2014;73:1935–41.
[43] Souza RB, Stehling C, Wyman BT, Hellio Le Graverand MO, Li X, Link TM, et al.
The effects of acute loading on T1rho and T2 relaxation times of tibiofemoral
articular cartilage. Osteoarthritis Cartilage 2010;18:1557–63.
 D. Hayashi et al. / Annals of Physical and Rehabilitation Medicine 59 (2016) 161–169
[44] Hovis KK, Stehling C, Souza RB, Haughom BD, Baum T, Nevitt M, et al. Physical
activity is associated with magnetic resonance imaging-based knee cartilage
T2 measurements in asymptomatic subjects with and those without osteoar-
thritis risk factors. Arthritis Rheum 2011;63:2248–56.
[45] Anandacoomarasamy A, Leibman S, Smith G, Caterson I, Giuffre B, Fransen M,
et al. Weight loss in obese people has structure-modifying effects on
medial but not on lateral knee articular cartilage. Ann Rheum Dis 2012;71:
26–32.
[46] Newbould RD, Miller SR, Toms LD, Swann P, Tielbeek JA, Gold GE, et al. T2*
measurement of the knee articular cartilage in osteoarthritis at 3T. J Magn
Reson Imaging 2012;35:1422–9.
[47] Wiliams A, Qian Y, Golla S, Chu CR. UTE-T2* mapping detects sub-clinical
meniscus injury after anterior cruciate ligament tear. Osteoarthritis Cartilage
2012;20:486–94.
[48] Madelin G, Babb J, Xia D, Chang G, Krasnokutsky S, Abramson SB, et al. Articular
cartilage: evaluation with fluid-suppressed 7.0T sodium MR imaging in sub-
jects with and subjects without osteoarthritis. Radiology 2013;268:481–91.
[49] Keen HI, Conaghan PG. Ultrasonography in osteoarthritis. Radiol Clin N Am
2009;47:581–94.
[50] Keen HI, Lavie F, Wakefield RJ, D’Agostino MA, Hammer HB, Hensor E, et al. The
development of a preliminary ultrasonographic scoring system for features of
hand osteoarthritis. Ann Rheum Dis 2008;67:651–5.
[51] Kortekaas MC, Kwok WY, Reijnierse M, Huizinga TW, Kloppenburg M. Osteo-
phytes and joint space narrowing are independently associated with pain in
finger joints in hand osteoarthritis. Ann Rheum Dis 2011;70:1835–7.
[52] Klauser AS, Faschingbauer R, Kupferthaler K, Feuchnter G, Wick MC, Jaschke
WR, et al. Sonographic criteria for therapy follow-up in the course of ultra-
sound-guided intra-articular injections of hyaluronic acid in hand osteoar-
thritis. Eur J Radiol 2012;81:1607–11.
169
[53] Conaghan PG, D’Agostino MA, Le Bars M, Baron G, Schmidely N, Wakefield R,
et al. Clinical and ultrasonographic predictors of joint replacement for knee
osteoarthritis: results from a large, 3-year, prospective EULAR study. Ann
Rheum Dis 2010;69:644–7.
[54] Wu PT, Shao CJ, Wu KC, Wu TT, Chern TC, Kuo LC, et al. Pain in patients with
equal radiographic grades of osteoarthritis in both knees: the value of gray
scale ultrasound. Osteoarthritis Cartilage 2012;20:1507–13.
[55] Nogueira-Barbosa MH, Gregio-Junior E, Lorenzato MM, Guermazi A, Roemer
FW, Chagas-Neto FA, et al. Ultrasound assessment of medial meniscal extru-
sion: a validation study using MRI as reference standard. AJR Am J Roentgenol
2015;204:584–8.
[56] Nakamura H, Masuko K, Yudoh K, Kato T, Nishioka K, Sugihara T, et al. Positron
emission tomography with 18F-FDG in osteoarthritic knee. Osteoarthritis
Cartilage 2007;15:673–81.
[57] Maas O, Joseph GB, Sommer G, Wild D, Kretzschmar M. Association between
cartilage degeneration and subchondral bone remodeling in patients with
knee osteoarthritis comparing MRI and (99 m)Tc-DPD-SPECT/CT. Osteoarthri-
tis Cartilage 2015;23:1713–20.
[58] Kalichman L, Li L, Kim DH, Guermazi A, Berkin V, O’Donnell CJ, et al. Facet joint
osteoarthritis and low back pain in the community-based population. Spine
2008;33:2560–5.
[59] Omoumi P, Michoux N, Thienpont E, Roemer FW, Vande Berg BC. Anatomical
distribution of areas of preserved cartilage in advanced femorotibial osteoar-
thritis using CT arthrography (Part 1). Osteoarthritis Cartilage 2015;23(1):
83–7.
[60] Omoumi P, Michoux N, Roemer FW, Thienpont E, Vande Berg BC. Cartilage
thickness at the posterior medial femoral condyle is increased in femorotibial
knee osteoarthritis: a cross-sectional CT arthrography study (Part 2). Osteo-
arthritis Cartilage 2015;23(2):224–31.