Best Practice & Research Clinical Rheumatology practice

Best Practice & Research Clinical Rheumatology 27 (2013) 249–269
Contents lists available at SciVerse ScienceDirect
Best Practice & Research Clinical
journal homepage:
How to interpret plain radiographs in clinical
Dr Andrew K. Brown, MBChB, MRCP, FHEA, PhD a, b, *
Senior Lecturer in Medical Education & Rheumatology, Hull York Medical School, University of York,
United Kingdom
Consultant Rheumatologist, York Teaching Hospital, NHS Foundation Trust, York, United Kingdom
a b s t r a c t
Plain radiographs
Conventional radiography
Rheumatoid arthritis
Ankylosing spondylitis
Psoriatic arthropathy
Calcium pyrophosphate deposition
In this article I will consider the basic principles of requesting,
acquiring, interpreting and reporting plain radiographs of joints,
including assessment of the distribution of joint abnormalities,
and specific pathological changes that may occur in bone, cartilage
and soft tissues. I will then move on to a more specific discussion
of the major arthropathies and the role of radiographs in the
diagnosis and assessment in each condition as well as reviewing
the combined abnormalities that may be visible on radiographs
and how these relate to underlying pathological processes.
Ó 2013 Elsevier Ltd. All rights reserved.
Conventional radiography (plain radiographs or X-rays) has been used by physicians to provide
additional information in patients with musculoskeletal symptoms for over a century. Despite advances in other imaging technologies such as ultrasonography (US), magnetic resonance imaging
(MRI), computed tomography (CT) and nuclear scintigraphy, including single photon emission
computed tomography (SPECT) and positron emission tomography (PET), plain radiographs remain an
important and widely used first-line investigation by health professionals in primary and secondary
This chapter will discuss the application of plain radiographs in the investigation and management
of patients with arthritis.
* Medical Education & Rheumatology, Hull York Medical School, University of York, York YO10 5DD, United Kingdom.
Tel.: þ44 1094 726308.
E-mail address: [email protected]
1521-6942/$ – see front matter Ó 2013 Elsevier Ltd. All rights reserved.
A.K. Brown / Best Practice & Research Clinical Rheumatology 27 (2013) 249–269
Advantages and disadvantages of plain radiography
It is worth first considering some of the potential advantages and disadvantages of conventional
One of the main advantages is that it is probably the most readily available and least expensive
imaging modality. This means that a wide range of health professionals are able to readily access these
data and almost all local patient populations are within a short distance of being able to have an X-ray
performed. All medical professionals have received at least some basic training in X-ray interpretation
and so this immediate access to information can generally be used by the requesting physician to aid
diagnosis and management in a time efficient fashion.
A wide range of pathological changes in bones, joints and cartilage can be evaluated with conventional
radiography. However, it is also important to consider that many soft tissue structures, including the
synovium, are not well visualised with plain radiographs and so X-rays cannot necessarily be relied upon
to provide the most sensitive information for diagnostic or therapeutic decision making in patients with
inflammatory arthritis. Other modalities such as US or MRI have a number of advantages in this context.
In addition, it is important to recognise that a radiograph effectively provides a two-dimensional
picture obtained from one slice through a three-dimensional structure. Therefore, the plane of
assessment is crucial in order to provide as much information as possible, as there may be a danger of
missing abnormalities if they are not caught in the path of the X-ray beam. For example, it is for this
reason that plain radiography may provide less sensitive information compared with topographical
techniques in the assessment of bone erosions in rheumatoid arthritis (RA).
Performing and receiving a radiographic assessment is a well-regulated process and it is generally
considered safe, although it does involve ionising radiation and the amount of radiation exposure
varies dependent on the structure being visualised. For example an X-ray of the pelvis involves
considerably more radiation exposure than an examination of a single limb joint, and the number of
views and images obtained are proportional to the amount of radiation received.
Data are generally regarded as reproducible as there are widely recognised standards for the
acquisition of images and increasingly there are also validated conventions and scoring systems for
interpreting and reporting various pathological processes, all of which are aimed at improving the
reliability and reproducibility of radiographic acquisition and interpretation. This is important as radiographs are often used for repeated serial evaluation and follow-up, for example, in the assessment
of progression of joint damage in RA. However, local variations remain relatively common.
Basic principles of requesting plain radiographs of joints
Which structures?
Before requesting the test, it is important to consider the capabilities of radiographs, that is, what
structures and pathology may be visualised using this modality and whether an alternative imaging
technique may provide more useful information. For example, radiographs can be used to evaluate
changes in joint and cartilage surfaces, bone pathology, joint space narrowing, fracture, subluxation
and dislocation. However, soft tissue structures such as tendons, ligaments and synovium are not well
visualised and alternative imaging may provide more sensitive and specific information. Often X-rays
are performed as a standard baseline imaging investigation but discussion with a musculoskeletal
radiologist may help to select the most appropriate subsequent test if additional information is still
required. Correlation with clinical findings and information from other investigation and imaging
results is always important and radiographs should not be interpreted in isolation.
Which joints?
In most cases, the choice of site of radiograph will be strongly influenced by a prior clinical
assessment, and an X-ray requested of the patient’s symptomatic joint or joints. Exceptions to this may
include a patient with RA where X-rays of both hands and feet may be used to evaluate extent of
disease or structural damage which may be repeated serially over time, or a skeletal survey in an
A.K. Brown / Best Practice & Research Clinical Rheumatology 27 (2013) 249–269
oncology setting, looking for evidence of sub-clinical malignant disease, such as myeloma. Nevertheless, a fundamental principle of requesting any imaging investigation is that any findings need to be
interpreted in clinical context, and it is therefore imperative that any X-ray request is accompanied by
comprehensive clinical information.
Which views?
Consideration should also be applied as to which views or plains are requested of the joint. Usually
it is appropriate to obtain two views at 90 apart, typically in antero-posterior (AP) and lateral or
oblique plains. This is particularly important in the setting of trauma but may be less crucial in the
assessment of arthropathy. There are a number of established techniques and protocols which have
been developed to provide optimal imaging of most joints and for particular indications. However, it is
important to remember that these only represent a two-dimensional ‘slice’ through a joint and so are
likely to be less sensitive than a three-dimensional or topographical assessment. In addition, specialist
views have been developed for particular indications. For example, a ‘ball-catchers’ view can be helpful
to visualise more of the cartilage surfaces in the finger joints and a ‘skyline view’ may be useful for
more accurate assessment of the patella-femoral joint.
Compare sides and review previous radiographs
It can often be helpful to compare findings in a symptomatic joint with those in a contralateral
asymptomatic joint. For example in a patient with osteoarthritis (OA), an AP X-ray of the pelvis allows
some comparison between both hip joints or in a patient with RA, an X-ray of both hands allows
comparison between finger, hand and wrist joints. If there is an old X-ray of the area of interest,
reviewing and comparing these images can also help to improve diagnostic certainty and evaluate any
progression over time.
Patient position
The position of the patient can also be important. For example, requesting weight-bearing views can
be much more informative in evaluating cartilage loss in the knees of a patient with OA or in providing
additional information concerning biomechanical changes in the feet.
Correlation with clinical and other imaging findings
This is an important principle, as mentioned earlier. Any radiographic findings need to be interpreted in
clinical context. Additionally further imaging may be required to provide more specific information. Discussion with a radiologist can often be helpful and a regular musculoskeletal radiology meeting can be a
useful forum for specialist clinicians and radiologists to formally discuss and review more challenging cases.
Important safety considerations
It is important to remember that performing a plain radiograph exposes a patient to ionising radiation. This is particularly important in women of child-bearing age. Radiographs of deeper structures,
such as the lumbar spine or pelvis, subject the patient to greater exposure than more superficial
structures. It is important to be able to justify any radiation exposure on the basis of potential risk and
benefit. The Department of Health Policy [1], (IR (ME)R 2000) covers this aspect in detail (
Basic principles of examining and reporting plain radiographs of joints
Most clinicians will have a straightforward strategy for reviewing and interpreting skeletal X-rays,
and this section aims to reprise some basic principles supplemented with the author’s experience and
present some ideas that may help further develop a systematic approach.
A.K. Brown / Best Practice & Research Clinical Rheumatology 27 (2013) 249–269
Clearly a knowledge and understanding of anatomy and pathological effects on bone, cartilage,
joints and soft tissues is crucial. Also it is important to be able to describe any obvious abnormality
using simple descriptive language, words and phrases, for example, name the bone and describe the
location and site of any abnormality, for example, right or left; proximal, middle or distal; head, neck or
shaft and cortex or medulla.
One should start by making sure that the radiographic and patient demographic data correspond and
the correct image is being viewed. Is it the correct date; is it the right or left side of the body; has the
region of interest been included and is positioning optimal and is there more than one view to assess?
It may be that there is an obvious abnormality visible on the radiograph, in which case I would
probably move straight on to describing and interpreting this. If not (and it is probably good practice to
still do this, in case there is an additional, perhaps less obvious abnormality present), a coordinated
systematic approach of formally looking at specific structures can be useful. This mental checklist
should ensure that the likelihood of missing any abnormalities is reduced.
1. Consider bony alignment – are there any changes that may suggest fracture or dislocation?
2. Consider bone cortices – follow the outline of each bone as any breach in the cortex may indicate a
fracture or arthropathy.
3. Consider bone texture – normally one expects to see a trabecular pattern within the substance of
bones but any distortion of this may indicate pathology.
4. Joint space – a careful look at the joint space may demonstrate changes such as joint space narrowing due to cartilage loss or calcification of the cartilage (chondrocalcinosis), or new bone formation, for example, osteophytes.
5. Soft tissues – changes in the soft tissue densities adjacent to a joint may indicate a joint effusion.
It is important to also review adjacent non-musculoskeletal structures looking for any pathological
Remember to look at all views, compare both sides and adjacent joints, review any previous images,
consider clinical findings and correlate with other imaging and test results.
Ask for a second opinion if any uncertainty, particularly a musculoskeletal radiologist, or consider
presenting the case at a multidisciplinary radiology conference.
A summary of the important factors to consider when interpreting a musculoskeletal radiograph is
provided in Table 1.
Radiographic abnormalities in bone, cartilage and soft tissues
In each of the major arthropathies, different pathological processes can cause changes to bone,
cartilage and soft tissue structures which may be demonstrable on radiographs. Whilst some of these
pathological changes may be specific, one should recognise an overlap across different types of arthritis.
In bone, changes include osteopenia, erosion, osteophyte and new bone formation, subchondral
sclerosis and cyst development. There may be cartilage loss and calcification (chondrocalcinosis). Soft
tissue swelling, calcification and deformity may be visible. Many of these changes may be in evidence
on radiographs of affected joints in the different forms of arthritis (Table 2).
Table 1
How to interpret a musculoskeletal radiograph – checklist.
Check patient demographic details
Is the image quality satisfactory? Are viewing angles optimal?
Describe the obvious abnormality using simple descriptive language e.g. name the bone and describe the location and
site of any abnormality
Use a systematic approach - start with the bones (alignment, cortices, texture) then joints, cartilage, soft tissues
Look at all views
Compare both sides / adjacent joints / previous images
Consider clinical findings
Correlate with other imaging
Ask for a second opinion – discuss case and clinical context with a specialist musculoskeletal radiologist
A.K. Brown / Best Practice & Research Clinical Rheumatology 27 (2013) 249–269
Table 2
Radiographic pathology in bone, cartilage and soft tissues in each major arthropathy.
Erosions adjacent
to the joint
New bone
New bone
Erosion away
from the joint
Joint space loss
(usually uniform)
Joint space loss
(usually asymmetric)
Joint space loss
Soft tissue
Soft tissue swelling
Bony remodelling
Soft tissue
Soft tissue
RA – rheumatoid arthritis; OA – osteoarthritis; PsA – psoriatric arthritis; AS – ankylosing spondylitis; CPPD – calcium pyrophosphate crystal deposition.
Assessment of distribution of joint abnormalities
When assessing a patient with arthropathy, it is often useful to consider not only which joints are
affected but also the pattern of distribution of joint involvement (Table 3). This can provide helpful
information not only in the initial stages of diagnosis but also when thinking about prioritising which
joints to assess during subsequent follow-up. These typical distributions of joint abnormalities are also
Table 3
The typical distribution joint involvement in the more common types of arthritis.
Type of arthropathy
Distribution of pathology
Rheumatoid arthritis
Typically an inflammatory polyarthritis affecting the small joints of the hands, feet
and wrists in a symmetrical distribution.
A mono, oligo or polyarthritis typically affecting DIP and PIP joints, first CMC joint,
axial skeleton and large weight bearing joints e.g. hips and knees.
Usually one of five sub-types:
1. predominant DIP joint involvement;
2. asymmetrical mono or oligoarthritis usually involving the knee and
small peripheral joints;
3. symmetrical peripheral polyarthritis resembling RA;
4. axial spondyloarthropathy/spondylitis;
5. arthritis mutilans associated with destruction, osteolysis and telescoping
of the fingers.
Usually involves the axial skeleton (sacroiliac joints and spine) (axial spondylitis).
May have asymmetrical involvement of medium and large joints particularly
shoulders and hips (peripheral spondylitis).
Inflammation at sites of bone insertion of tendons and ligaments is common
(enthesitis) e.g. iliac crests, gluteal and tibial tuborosities and heels.
Typically asymmetrical mono or oligoarthritis of weight bearing lower limb joints,
but other joints may be involved and may present as a polyarthritis.
Usually a monoarthritis, most commonly first MTP joint, then mid-foot, ankle
and knee; lower limb > upper limb.
Usually a monoarthritis, most commonly knee then wrist, shoulder, ankle
and elbow joints.
Usually an oligoarthritis, most commonly knee then wrist, shoulder, elbow, hip,
midtarsal and MCP joints.
Psoriatic arthritis
Ankylosing spondylitis
Reactive arthritis
Acute calcium pyrophosphate
arthritis “pseudogout”
OA with CPPD (previously
often called chronic
pyrophosphate arthropathy)
Septic arthritis
Usually a monoarthritis most commonly affecting lower limb large/medium joints
e.g. knee or less commonly hip although upper limb joints may be affected.
May occasionally present as a polyarthritis if immunocompromised or underlying
arthropathy such as RA.
DIP – distal interphalangeal; PIP – proximal interphalangeal; CMC – carpometacarpal; MCP – metacarpophalangeal; MTP –
metatarsophalangeal; CPPD – calcium pyrophosphate crystal deposition; OA – osteoarthritis; RA – rheumatoid arthritis.
A.K. Brown / Best Practice & Research Clinical Rheumatology 27 (2013) 249–269
important to take into account when choosing which joints to evaluate with plain radiographs and also
when interpreting these images.
Combined radiographic abnormalities that may occur with each major arthropathy
Rheumatoid arthritis
Plain radiography remains the gold standard for the assessment of structural joint damage in RA
even though this may not necessarily be the most sensitive imaging investigation in this setting.
Nevertheless, radiographs remain very important in the evaluation of these patients and have been
historically used as a primary outcome measure in RA. Indeed, much of the evidence we commonly
apply in the diagnosis and management of our RA patients continues to be based on longitudinal
radiographic information [7].
When a patient with inflammatory arthritis is initially assessed in the rheumatology outpatient
clinic, as well as a clinical, biochemical and immunological investigations, it is usual to perform plain
radiographs of the hands and feet. This is supported by most national rheumatology guidelines [8].
Even if symptoms are predominantly in the hands, it is important not to neglect the feet and perform a
radiographic assessment of the hands, wrists and feet, and any other affected joints at the physician’s
discretion, as erosive changes may be present in the feet and not the hands in early disease [9]. This
may improve the diagnostic utility of radiographs in the baseline evaluation of patients with early
disease. Large joint radiographs of RA patients have been less frequently studied; however, studies
including large joints of patients with established disease indicate good correlation between radiographic erosions of large and small joints [10]. This information can be useful to not only help establish
a diagnosis but also determine the extent of disease and help ascertain risk of progression and prognosis. For example, the presence of radiographic erosions often confers more severe disease and worse
prognosis [2] and a number of studies have demonstrated a correlation between joint damage seen on
radiographs and disability in long-standing RA, although this link is perhaps less strong in patients
with early disease [3].
It is important to note that X-rays are often normal at onset and during the early stages of the
condition and the more characteristic radiographic features may only develop in more established
disease. Studies have estimated that less than half of new RA patients attending their first rheumatology clinic visit may have visible radiographic erosions and in the remainder of patients X-rays are
often normal [12,13]. Moreover, a third of new RA patients may not develop radiographic erosions
within the first 2 years of disease onset [11]. It is therefore important to also evaluate other nonradiographic markers of disease severity when planning individual patient management. This has
led to some authors questioning the utility of radiographs in the assessment of early RA, as bony
changes visible on X-ray often lag behind clinically detectable joint and soft tissue inflammation. In
addition, studies comparing radiographs with other imaging techniques such as US, MRI and CT have
all confirmed the reduced sensitivity of X-rays at detecting early erosive changes [14–16].
Other studies suggest that the majority of patients will have developed some radiographic changes
within 2 years of diagnosis [17,18] and most national recommendations support the value of a baseline
radiographic assessment although caution should be applied when interpreting the significance of
normal X-rays in this situation.
Radiographs may therefore perhaps be more usefully applied in the serial assessment of joints
affected by RA, looking for evidence of progression of joint damage over time. Nevertheless, most
specialists would suggest that radiographs of hands and feet should be performed at baseline and every
6–12 months in early RA and perhaps every 1–2 years in more established disease.
This is reflected in recent modifications to the RA classification criteria. Previous criteria, which
included radiographic changes [4], were mainly applied to clinical and epidemiological research
studies but have only been shown to be sensitive and specific for the diagnosis of active, established RA.
Their sensitivity may be low in early RA, particularly <12 weeks duration [5], as the classical features of
RA such as radiographic erosions and rheumatoid nodules, which are included in these criteria, are
often absent at disease onset. In response to this, a collaborative initiative undertaken by American
College of Rheumatology (ACR) and European League Against Rheumatism (EULAR) was launched
A.K. Brown / Best Practice & Research Clinical Rheumatology 27 (2013) 249–269
Fig. 1. Early rheumatoid arthritis. Bone erosions are visible in the left index finger MCP joint and left middle finger PIP joint.
which has resulted in the publication of the 2010 Rheumatoid Arthritis Classification [6], which
focusses on features at an earlier stage of disease, rather than defining the disease by late stage features, and does not include radiographic changes.
A range of pathological abnormalities may be visualised radiographically in RA. These include soft
tissue swelling, periarticular (juxta-articular) osteopenia, erosions, joint space narrowing (usually
uniform) and deformity such as subluxation and dislocation (Figs. 1 and 2). It is important to remember
that radiographs only provide limited information on soft tissue structures and US or MRI are the
Fig. 2. Severe established rheumatoid arthritis. There is advanced destructive arthropathy with erosive disease predominantly
affecting the MCP and wrist joints bilaterally. This is most marked on the right with marked MCP joint damage, subluxation and
deformity with ulnar deviation. Joint space and cartilage loss is demonstrated throughout. There has been severe destruction and
ankylosis of the carpal bones bilaterally. There is pronounced osteopenia throughout.
A.K. Brown / Best Practice & Research Clinical Rheumatology 27 (2013) 249–269
imaging modalities of choice to directly visualise these structures and provide important objective
information on pathological changes such as synovitis and tenosynovitis. For this purpose, radiographic assessment relies on the interpretation of soft tissue shadows corresponding to, for example,
elevation of the fat pad in the elbow joint or suprapatellar pouch of the knee, which may suggest a joint
effusion or synovitis, and is therefore not necessarily sensitive or specific for assessing inflammation.
These pathological abnormalities can be observed in a range of joints with the distribution largely
correlating with the established clinical pattern of a symmetrical peripheral joint polyarthritis, characteristic of RA. As such it is usually the metacarpophalangeal (MCP) and proximal interphalangeal
(PIP) joints in the fingers, metatarsophalangeal (MTP) joints in the forefeet and all compartments of the
wrists that are most commonly involved. In addition, joints in the midfoot and hindfoot, knees, glenohumeral joint at the shoulder, the elbow and cervical spine can also be affected.
The most common first radiographic sign is often periarticular osteopenia as bone density is
reduced adjacent to the joint as a result of local synovial inflammation. The bone may appear less dense
(a darker shade on the radiograph) around the articular surfaces, although this is not necessarily a
specific radiographic sign of RA and can occur in other conditions.
When synovial inflammation is not controlled, through either delayed diagnosis or lack of response
to therapy, then cartilage and bone destruction can occur. The inflamed synovium slowly invades
adjacent structures causing damage and destruction to the cartilage and bone leading to joint space
narrowing and bone erosion that can be seen on radiographs (Fig. 1). The joint space narrowing in RA
tends to be uniform and concentric reflecting the generalised nature of the synovial inflammation
within the joint, whereas in other conditions such as OA, it may be more uneven and asymmetric. The
erosions in RA tend to be periarticular and are often described as marginal erosions as they are close to
the joint and reflect the site of pathology. The most likely sites to visualise early erosive changes in RA
include the radial aspects of the second and third MCP joints in the hand, the ulnar styloid at the wrist
and the lateral aspect of the fifth MTP joint.
If the disease remains untreated, progression inevitably occurs resulting in more extensive erosive
joint damage and destruction within the joint. There may also be progressive distension of the joint
capsule that increases joint instability, which causes biomechanical compromise and may also accelerate joint damage. These factors contribute to the development of joint subluxation and ankylosis and
characteristic deformities such as ulnar deviation and subluxation resulting in boutonnière and swan
neck deformities of the finger joints (Fig. 2).
It is always important to consider the cervical spine as part of a radiographic assessment in patients
with RA. In cases of established RA, up to 80% of patients may develop changes in this region, usually at
the atlanto-axial joint as a result of progressive erosion and bone damage reflecting ongoing, uncontrolled inflammation and pannus formation. In extreme cases, this joint can become subluxed in either
an anterior, lateral or vertical direction with potentially serious consequences of spinal cord or
brainstem compression. Radiographs can be used to assess the alignment of this region using AP and
lateral views with comparison of images with the neck held in a forward flexion and neutral or
extended position (Fig. 3). If the anterior atlanto-dental interval (AADI) is >3 mm, then atlanto-axial
subluxation should be suspected and further imaging usually in the form of an MRI scan should be
performed for more accurate evaluation of inflammation and erosive damage and comprehensive
assessment of bony, soft tissue and neurological structures [19].
Progression of structural damage to joints is commonly used as an outcome measure in RA. This is
most commonly measured by applying scoring systems to assess radiographs of the hands, wrists and
feet. Several validated scoring systems have been developed for radiographic evaluation of RA pathological features. The best known of these are the Sharp and Larson scores. They tend to be rather
complex and involve summating scores for erosions and joint space narrowing in the hands, wrists and
feet, perhaps limiting their utility in routine clinical practice, but they are nevertheless important in
research studies, particularly as an outcome measure for measuring the efficacy of therapies and
assessing longitudinal response to treatment. Extensive research has confirmed both their validity and
reliability amongst appropriately trained practitioners for the measurement of disease severity and
particularly progression of bone and joint destruction over time [20]. The radiographic initial score at
the time of diagnosis has been shown to be a reliable predictor of future radiographic structural
damage [21].
A.K. Brown / Best Practice & Research Clinical Rheumatology 27 (2013) 249–269
Fig. 3. Atlanto-axial subluxation in rheumatoid arthritis. Anterior atlanto-axial subluxation is present on neck flexion as the anterior
atlanto-dental interval (AADI) (arrow) measures 10 mm. Note erosive changes in the odontoid peg. AADI is measured from the
postero-inferior margin of the anterior arch of the atlas to the anterior surface of the odontoid; greater than 3 mm is considered
abnormal; 3–6 mm indicates early instability; greater than 9 mm is regarded by some authors as an indication for surgical stabilisation although should be correlated with clinical and MRI findings [19].
In summary, baseline radiographs may be useful in some patients to identify features of RA such as
periarticular osteopenia, joint space narrowing or erosions. Repeated radiographic assessment is an
established measure of disease progression, patient outcome and response to treatment. In addition,
consideration should be given to radiographic assessment of other joints particularly the atlanto-axial
region of the cervical spine in patients with established and particularly severe RA, to assess for evidence of bone damage and subluxation.
Plain radiographs remain the standard investigation in the assessment and diagnosis of OA. In the
clinical setting, imaging is usually targeted to the symptomatic joint but often other joints can be
affected by OA but not necessarily the symptomatic joint. The opposite situation may also be true,
meaning that clinical correlation is always important.
The typical distribution involves the large lower limb weight-bearing joints of the knee and hip,
further up the kinetic chain in the lumbar and cervical spine, and peripheral joints including the thumb
bases (especially dominant hand), proximal and distal interphalangeal joints of the fingers and first
metatarsophalangeal joints in the feet. Any joint articular surface that has been subject to previous
trauma or fracture may be at risk of developing premature OA.
Radiographs can be used to inform a diagnosis of OA, determine disease severity, in monitoring, for
example, post joint replacement, and to evaluate structural progression. However, sensitivity and
specificity may be less for detecting early changes of OA. In addition, the optimal frequency of repeat
radiographs to inform clinical management is not well defined.
As with other types of arthritis, due consideration is needed as to the plain of assessment (e.g., a
‘sky-line’ view to assess the patella-femoral joint in the knee), positioning (e.g., X-ray the hip joint in a
standing position with 15–20 of internal rotation) and correlation with clinical and other imaging
findings. Post X-ray processing techniques such as digital image analysis and magnification techniques
A.K. Brown / Best Practice & Research Clinical Rheumatology 27 (2013) 249–269
can help. Standardised and consistently applied image acquisition techniques are clearly important to
ensure reliable and reproducible images particularly to allow meaningful comparison over time, for
example, post joint replacement surveillance and in the assessment of structural progression.
The radiographic hallmark of OA is loss of joint space reflecting degeneration and thinning of
articular cartilage. Meniscal cartilage lesions and cartilage extrusion may also contribute. Joint space
narrowing is more likely to be asymmetrical reflecting localised cartilage damage rather than the more
usual symmetrical appearances in inflammatory arthropathies such as RA. Weight-bearing X-rays may
be useful to more accurately assess the degree of joint space narrowing, particularly in joints such as the
knee and hip, to ensure that the articular surfaces are in direct contact at the time the X-ray picture is
taken. The subchondral bone can be affected by excess load-bearing forces leading to cyst formation. In
response to cartilage loss, a pathophysiological repair process occurs with new bone formation producing osteophytes at the joint margin, and thickening and sclerosis of the subchondral bone, visible as
increased ‘whitening’ on the radiograph (Fig. 4). Eventually collapse, subluxation and bone remodelling
may occur (Fig. 5) although ankylosis is uncommon. Abnormal alignment may result, for example, varus
or valgus deformities affecting the knee. Erosions may be seen but unlike RA or gout these are usually
centrally located within the joint, characteristically in the distal interphalangeal joints of the fingers
where they have been described as a ‘seagull wing’ appearance (‘erosive’ OA). Particularly in the early
phase of the disease, inflammation may contribute perhaps provoked by products of cartilage damage,
causing soft tissue swelling which may be visible on the radiograph. It should be noted that other arthropathies may co-exist, particularly calcium pyrophosphate crystal deposition (CPPD), producing
additional radiographic features including chondrocalcinosis, and that OA can be a secondary reaction
to cartilage damage produced by inflammatory joint disease such as RA. Demineralisation (osteopenia)
is usually absent and would be much more in keeping with an inflammatory arthropathy such as RA.
A number of scoring systems have been developed in order to attempt to provide more objective
measures as to the severity of OA. Probably one of the oldest and most well-known, but still widely
used, systems is the Kellgran and Lawrence system. This may be used to classify the severity of OA
according to a five-point scale:
Grade 0: Normal.
Grade I: Doubtful: that is, possible narrowing of the joint space and osteophytes.
Fig. 4. Osteoarthritis of the hands. Joint space loss, subchondral cysts and osteophytes are present particularly at the left first CMC
joint. Similar changes with the addition of subchondral sclerosis are also present in a number of DIP and PIP joints with subluxation
and deformity of the right index finger DIP joint.
A.K. Brown / Best Practice & Research Clinical Rheumatology 27 (2013) 249–269
Fig. 5. Osteoarthritis of the hips. Advanced osteoarthritis is present in the left hip with marked joint space loss, sclerosis, osteophyte
formation with remodelling and superior migration of the femoral head. Significant osteoarthritis is also present in the right hip
with joint space loss, subchondral cysts and osteophytes. Note new bone formation around entheseal attachments at the greater
trochanters and around the pelvis.
Grade II: Minimal: that is, small osteophytes, mild narrowing of the joint space.
Grade III: Moderate: that is, multiple, moderately sized osteophytes, definite joint space narrowing, some sclerotic areas and possible deformity of bone contour.
Grade IV: Severe: that is, multiple large osteophytes, severe joint space narrowing, marked sclerosis and definite deformity of bone contour.
Adapted from Ref. [36]
In daily clinical practice a semi-quantitative scale is often employed describing pathological
changes, particularly joint space narrowing, as minimal, moderate or severe.
Quantitative measurement of joint space width, usually in millimetres, is often used in research
studies to evaluate OA structural progression and effects of therapy on structure modification and
remains the optimal recognised technique for this purpose. This is applied most commonly in the knee
and also has been used in the hip and hand joints. However, it can be a challenging outcome measure
with which to demonstrate any change over time and may not be a reliable surrogate for cartilage loss
and subject to other confounding variables [37].
Psoriatic arthritis
Plain radiographs are also an important imaging modality in the assessment of patients with
psoriatic arthropathy (PsA) although the greater range of differing manifestations of this disease means
that varying radiographic patterns of joint and soft tissue involvement can be observed.
A range of pathological abnormalities may be visualised radiographically in PsA. Similar to RA, these
may include soft tissue swelling, joint space narrowing and erosions but differentiating features
include in particular proliferative new bone formation as well as periostitis, ankylosis and osteolysis.
These pathological abnormalities can be observed in a range of synovial joints with changes also
affecting the axial skeleton and fibrocartilaginous joints such as the sacroiliac joints and entheseal
attachments of tendons and ligaments. The distribution can be much more variable with up to five
recognised patterns including predominant distal interphalangeal joint involvement; asymmetrical
mono- or oligoarthritis usually involving the knee and small peripheral joints; a symmetrical peripheral polyarthritis resembling RA; an axial spondyloarthropathy or spondylitis and finally arthritis
mutilans associated with destruction, osteolysis and telescoping of the fingers.
Up to a quarter of patients with PsA may have sacroiliitis and changes may be more extensive and
are more likely to be asymmetrical in PsA and reactive arthritis than in ankylosing spondylitis (AS) [22].
Spondylitis is also relatively common and often indistinguishable from AS, although in PsA
A.K. Brown / Best Practice & Research Clinical Rheumatology 27 (2013) 249–269
syndesmophytes may be less extensive and are less likely to form consecutive bridging between
adjacent vertebrae. Typical patterns of joint involvement in the hand include inflammatory changes in
the distal and proximal interphalangeal joints or a combination of diffuse soft tissue swelling, and
tenosynovitis and synovitis characteristic of dactylitis, which can also affect the toes (‘sausage-digits’)
[23]. Changes in the feet often involve the MTP and interphalangeal joints, particularly the interphalangeal joints of the great toe. Phalangeal tufts can also be affected. Not uncommonly the distribution of
affected joints can mimic RA with symmetrical synovial swelling of the small joints of the hands, feet
and wrists which can damage and destroy joints in a similar fashion to RA [24]. Indeed, it has been
suggested that PsA may be a more disabling and erosive condition than previously recognised [25]. The
feet are commonly affected in PsA. An inflammatory oligoarthritis can also occur with the involvement
of small and/or large joints in particular involving the ankles, knees and shoulders.
Like in RA, the expression of radiographic changes in early disease is often inconsistent and it is not
until the disease is established that the more characteristic radiographic changes may be seen. Up to
half of patients with early PsA may have evidence of radiographic bone damage within 2 years of
presentation [26]. However, there is characteristically a combination of bone destruction and new bone
formation. Erosions may be seen which can be indistinguishable from those seen in early RA in that
they are characteristically well defined and situated in a peri-articular location but are more likely to be
asymmetric. However, in contrast to RA, these changes are usually associated with proliferative new
bone formation which can give the erosion a speculated appearance particularly towards its margins
(Fig. 6). Other changes that can be seen may include soft tissue swelling due to dactylitis, periostitis
which can affect the bony shaft and spondylitis.
If the disease does not respond to treatment and progress, the bone destruction may worsen meaning
the erosions become more irregular and indistinct whilst new bone formation continues. This can result in
some characteristic deformities including ‘pencil-in-cup’ changes affecting the distal interphalangeal joints
with the middle phalanx taking on a pointed tip appearance through progressive erosion and osteolysis,
whereas the base of the distal phalanx expands and curves laterally as new bone is formed (Fig. 7).
Further bony proliferative change can occur as a result of ongoing entheseal inflammation. These
can be visible on radiographs and most commonly occur in sites such as the achilles tendon and plantar
fascia insertion sites at the calcaneum as well as at sites of tendinous and ligamentous attachments
around the pelvis.
Like the other major arthropathies, scoring methods have been developed in an attempt to quantify
changes more objectively. Examples include the modified Steinbrocker score, Ratingen method for PsA,
Sharp score (similar to RA) and van der Heijde modification of the Sharp method [32].
Fig. 6. Psoriatic arthropathy of the fingers. Early changes of psoriatic arthropathy affecting the interphalangeal joints of both thumbs
and DIP joints of all the fingers in the right hand and the left middle finger. There is evidence of joint space loss, erosions and new
bone formation.
A.K. Brown / Best Practice & Research Clinical Rheumatology 27 (2013) 249–269
Fig. 7. Psoriatic arthropathy of the toes. More advanced changes are present in the feet with advance erosive changes present
particular at the right great to PIP joint where there has been marked osteolysis and early “pencil and cup” deformity.
Ankylosing spondylitis
AS almost universally affects the sacro-iliac joints as well as the axial skeleton. Bone erosion, new
bone formation, ankylosis and enthesitis are common pathological features all of which may be
visualised on radiographs.
Radiographs remain an important tool in informing the diagnosis of AS. Usually the first radiographic changes of AS can be found in the sacroiliac joints as symmetrical sacroiliitis (Fig. 8), and such
findings have a high specificity for this condition. However, although up to 95% of AS patients may
Fig. 8. Ankylosing spondylitis associated sacroiliitis. Early sacroiliitis is demonstrated on this radiograph of the pelvis with loss of
clarity and sclerosis in the lower third of the sacroiliac joints, particularly affecting the iliac side of the right sacroiliac joint. Hip joint
appearances are normal.
A.K. Brown / Best Practice & Research Clinical Rheumatology 27 (2013) 249–269
develop radiographic changes, they may take up to 9 years to appear as X-rays are only able to detect
more chronic bony changes [33].
The first radiographic finding of sacroiliitis is usually seen in the synovial portion of the joint which
comprises the anterior and inferior one-half to two-thirds (the remainder of the joint is ligamentous).
Typically appearances are bilateral and symmetrical. The iliac aspect of this region is usually the best
place to look, as the articular cartilage tends to be thinner here. One should look for erosions at this site
although early changes can be rather subtle. Focus on the margins of the joint and look for loss of clarity
and definition of the white line of the bone cortex which can progress to give the appearance of a
widened joint space. Bony proliferation then tends to occur as part of the body’s repair response
producing sclerosis and increased whitening of the bone on the radiograph. As the disease process
progresses, the joint can eventually fuse or ankylose (Fig. 9) and as such lose its ability to move, which
can alter the load bearing and create more mechanical stresses in the adjacent structures. The ligaments holding together the posterior portion of the joint can also calcify, contributing to the sclerotic
radiographic appearances of this region, often obscuring the joint itself and making interpretation
more challenging.
The sensitivity of radiography to detect these changes can be improved by using a specific angulated
viewing position, such as the modified Ferguson view, whereby the patient lies on their back with
knees and hips flexed and the X-ray tube is centred at the L5-S1 level and then angled 25–30 towards
the head. One should usually not rely on a standard AP X-ray of the pelvis as the optimal method of
visualising the sacroiliac joints. However, a pelvic radiograph may be beneficial to evaluate other areas
which may be affected in cases of spondyloarthropathy such as the hips and pubic symphysis, looking
for erosions and new bone formation, as well as ligament and tendon attachment around the pelvis,
looking for changes of enthesopathy.
The diagnosis of AS is usually made by applying criteria that combine clinical and radiographic
findings. Amongst the most commonly used is the New York Criteria which requires demonstration of
either bilateral sacroiliitis at least grade 2 (minimal sacroiliitis: loss of definition of the joint margins,
minimal sclerosis, joint space narrowing and erosions) or unilateral sacroiliitis at least grade 3
(moderate sacroiliitis: definite sclerosis on both sides of the joint, erosions and loss of joint space) [31].
However, these changes may occur a few years after the development of characteristic symptoms of
inflammatory back pain so their diagnostic utility may not be optimal for early disease. This probably
relates to a number of factors including the difficulty in optimally imaging such an obliquely angled
joint with its twisting articular surfaces and complex anatomy, relatively poor inter- and intra-rater
reliability in image interpretation and the fact that radiographs are only able to detect established
bony changes in these joints and not active inflammation [30]. This relatively poor sensitivity has led to
recent modifications to the classification criteria for axial spondyloarthropathy which now allow
Fig. 9. Advanced ankylosing spondylitis. Advanced AS with ankylosis or fusion of the sacroiliac joints.
A.K. Brown / Best Practice & Research Clinical Rheumatology 27 (2013) 249–269
identification of sacroillitis by MRI [27]. MRI can identify both inflammation and bone erosions of the
sacroiliac joints and is more sensitive than conventional radiography in the detection of sacroiliitis and
may detect changes despite normal X-rays [29]. Some studies have estimated that the additional rate of
detection with contrast-enhanced MRI may be as much as 75% [28]. Precise MRI sequencing can help,
and in particular using fat suppression techniques such as STIR sequences can improve detection of
bone marrow oedema which correlates well with active inflammation. As such MRI may be a more
useful imaging tool in detecting early axial spondyloarthropathy.
Besides the sacroiliac joints, similar pathological changes of AS may be seen in the cervical, thoracic
and lumbar spine. Radiographs may demonstrate erosions, new bone proliferation, ligamentous
ossification as well as squaring and fusion of the vertebral bones.
The first changes usually appear at the thoracolumbar junction with small erosions and adjacent
new bone formation at the corners of the vertebral bodies. This produces a ‘shiny corner’ appearance
which is called a Romanus lesion. Bony proliferation usually dominates leading to further new bone
formation along the anterior aspect of the vertebral body which may cause loss of the normal concave
contour giving the vertebral body a squared-off appearance. As the condition progresses, ossification
and calcification can begin to affect the outer fibres of the annulus fibrosis of the intervertebral disk and
the longitudinal spinal ligaments forming syndesmophytes. These specific lesions can further develop
to form vertical bony spurs which may eventually form a bridge between adjacent vertebrae leading to
a so-called ‘bamboo spine’ appearance, which is characteristic of severely established AS (Fig. 10). The
facet joints can be similarly affected. New bone formation and ossification of other soft tissue structures
can also occur, particularly the posterior interspinous ligament which in severe cases can be visualised
as a solid radio-opaque white band in the midline. Similar to the sacro-iliac joints, ongoing inflammation can eventually result in portions of the spine becoming ankylosed resulting in loss of mobility
and functional impairment and disability. As a result of altered biomechanics and load bearing,
Fig. 10. Advanced ankylosing spondylitis. Advanced AS with bridging ostophytes affecting the thoracic and lumbar spine.
A.K. Brown / Best Practice & Research Clinical Rheumatology 27 (2013) 249–269
fractures may occur through an ankylosed disk or vertebral body creating a pseudo-arthrosis, often
referred to as an Andersson lesion.
These more chronic radiographic changes in the spine are not part of current diagnostic criteria for
AS because the presence of spinal changes in association with radiographically normal sacroiliac joints
is unusual.
The current ASAS recommendations suggest that X-rays of the cervical and lumbar spine should be
performed. Their utility is probably greatest in helping to determine extent of disease and also for
following longitudinal progression. Lateral views are usually the most sensitive ones and involve less
radiation exposure. ASAS recommendations suggest that although changes in the thoracic spine can be
seen relatively frequently, interpretation can be more challenging due to additional shadows projected
from overlying soft tissue and organ structures [34]. The frequency of repeat radiographs to assess
structural progression is uncertain but most authors seem to suggest no less than every 2 years, given
the usual slow rate of progression of these bony changes in AS.
A number of validated scoring systems have been developed for assessment of radiographic changes
of AS in the spine and include the modified Stoke Ankylosing Spondylitis Spinal Score (mSASSS) which
assesses 24 sites on the lateral cervical and lumbar spine for sclerosis, squaring or erosion; syndesmophytes and bony bridge formation. Other methods also exist and include the Bath Ankylosing
Spondylitis Radiological Index (BASRI) which also includes the sacroiliac joints and the Stoke Ankylosing Spondylitis Spinal Score (SASSS). Outcome Measures in Rheumatoid Arthritis Clinical Trials
(OMERACT) has selected the mSASSS as the preferred method of scoring structural damage in AS, as
this system is supported by data confirming good reproducibility and sensitivity to change [35].
As discussed above, diagnostic and classification criteria have evolved over time and authors often
now refer to this spectrum of diseases as spondyloarthritis, subdivided into axial spondyloarthritis
(with AS being the prototype disease) and peripheral spondyloarthritis [34]. Certainly, involvement of
joints and entheses of the peripheral skeleton is relatively common. The distribution of joint
involvement is more commonly asymmetrical, affecting medium and large joints particularly hips,
shoulders and knees as well as metatarsophalangeal joints. Pathological changes may be detected
radiographically. The hips are most frequently involved and changes are usually bilateral and symmetrical and the presence of hip disease may be a poor prognostic sign. Typically there is uniform
narrowing of the joint space (reflecting synovial inflammation) and osteophyte formation at the
junction of the femoral head and neck often forming a ‘collared’ appearance. With ongoing joint
damage the femoral head may migrate upwards or there may be intra-pelvic displacement of the
medial acetabular wall, referred to as ‘acetabular protrusio’. A third of patients may experience
shoulder involvement and again features tend to be bilateral and symmetrical. Typical radiographic
features include joint space narrowing, erosions, bony proliferation of ligament attachments and the
possibility of ultimately ankylosis. Inflammation at sites insertion of tendons and ligaments into bone
(enthesitis) is common are more difficult to demonstrate with radiographs until the point of new bone
and enthesophyte formation. Such changes may be seen in sites such as the iliac crests, gluteal and
tibial tuborosities and typical bony spurs at the achilles tendon and plantar fascia insertions in the
calcaneum may also be detected. Other imaging techniques such as US and MRI may be more sensitive
at detecting these entheseal changes at an earlier stage.
Crystal arthropathy
These are a range of arthropathies characterised by crystal deposition within joints and periarticular tissues. The most common types are gout and calcium pyrophosphate disease. The radiographic findings in both of these related conditions will be considered.
i. Gout
Gout is an inflammatory arthropathy associated with deposition of monosodium urate crystals in
joints, resulting from chronic elevation of tissue urate levels. It is a common condition, especially in
middle-aged and older men and is often associated with modifiable risk factors, for example, obesity,
excessive beer or spirit intake, hypertension, diuretic use and renal impairment. The diagnosis is
A.K. Brown / Best Practice & Research Clinical Rheumatology 27 (2013) 249–269
usually made clinically and is confirmed by demonstration of monosodium urate crystals (negatively
birefringent needles) on compensated polarised light microscopy of synovial fluid. Identification of
crystals is fundamental for making the correct diagnosis and excluding other causes such as sepsis.
Radiographs can be helpful in more chronic cases.
A typical initial presentation is an acute, severe, self-limiting inflammatory monoarthritis involving
the great toe metatarsophalangeal joint (podagra). It can also present as a monoarthrits in other joints,
oligoarthritis, polyarthritis or chemical cellulitis. Chronic recurrence is common, associated with tophi
formation and joint damage.
Radiographs can be used as part of the diagnostic process (although not included in the diagnostic
criteria) and to monitor progression. Radiographic changes are usually monoarticular and asymmetrical and are usually a legacy of recurrent disease. In acute gout, radiographs may be normal although
soft tissue swelling may be visible. In cases of chronic gout (recurrent acute episodes over time) a
variety of radiographic features may be visible. Often the earliest changes are bone erosions as
recurrent inflammation causes damage to the bony surfaces. The classic erosion appearance is of a
punched-out lesion with sclerotic margins and overhanging edges which is commonly located in the
para-articular area but may be located some distance away from the joint (note different site and
appearance to a rheumatoid erosion and the absence of periarticular osteoporosis and, unless severe,
maintenance of the joint space). However, recurrent attacks of acute inflammation may result in
progressive bony destruction with loss of joint space, erosive progression and deformity (Fig. 11). It is
also important to look at the soft tissues as soft tissue densities or calcified opacities may be apparent
caused by recurrent uric acid crystal deposition and formation of tophi.
ii. Calcium pyrophosphate crystal deposition
CPPD is characterised by deposition of calcium pyrophosphate crystals predominantly in articular
hyaline and fibrocartilage (chondrocalcinosis). There are familial and sporadic forms and the condition
may be associated with underlying metabolic disease. It is a common cause of acute inflammatory
monoarthritis in the elderly. It has a variety of clinical manifestations:
Fig. 11. Gout. This is advanced gouty arthropathy with marked erosive bone destruction and resultant deformity. Note the characteristic appearance and site of the erosion in the head of the first metatarsal bone when a slightly rotated plain is used. Tophi are
visible in the soft tissues between the first and second metatarsal bones.
A.K. Brown / Best Practice & Research Clinical Rheumatology 27 (2013) 249–269
asymptomatic incidental finding of chondrocalcinosis;
associated with acute inflammatory mono- or oligoarthritis (acute CPP crystal arthritis – ‘pseudogout’)
chronic arthritis – OA plus CPPD (previously termed chronic pyrophosphate arthropathy).
Like gout, the diagnosis is usually made clinically and is confirmed by demonstration of characteristic crystals (non-birefringent or weakly positively birefringent rods or rhomboids) on compensated polarised light microscopy of synovial fluid.
Typical distribution of joint involvement includes large and medium-sized joints, commonly knees
and also wrists, shoulders, ankles, elbows and others.
Plain radiographs are useful to demonstrate calcification and can sometimes aid diagnosis. This can
affect a variety of sites. Chondrocalcinosis is characteristic and can involve both fibrocartilage, for
example, knee menisci, wrist triangular fibrocartilage and symphysis pubis, and hyaline cartilage, for
example, knee, gleno-humeral joint and hip (Fig. 12). Calcification may also occur in the joint capsule
and synovium, for example, in the MCP and knee joints. There may also be calcification of soft tissue
structures particularly the entheses of achilles, triceps, obturator tendons and bursae in subacromial,
olecranon and retrocalcaneal sites. It should be remembered that chondrocalcinosis is a relatively
frequent finding in otherwise healthy people. In fact it may be present in up to a third of those aged 65–
75 with increasing incidence in older populations, the majority of whom are usually asymptomatic.
Chondrocalcinosis is also associated with other metabolic conditions including hyperparathyroidism,
haemochromatosis, hypophosphatasia and hypomagnesaemia.
In more chronic cases, structural changes may develop. These often resemble OA with cartilage loss,
sclerosis, cysts and osteophytes (Fig. 13). Features more specific to CPPD include more prominent
‘exuberant’ osteophytes and large subchondral cysts particularly at the knee and wrist joints which
contribute to the so-called ‘hypertrophic’ appearance characteristic of this condition. The involvement of
other joints less typically affected by OA, for example, MCP, gleno-humoral, ankle, mid-foot and radiocarpal (often associated with scaphoid-lunate dissociation), can also provide useful clues to this diagnosis. Smooth ‘pressure’ erosions may occur, for example, at the anterior distal femur, distal inferior radioulna joint and radio-carpal joint. The relatively rare variant of destructive pyrophosphate arthropathy may
cause marked cartilage and bone loss over a fairly short period of time, often resulting in changes
Fig. 12. Calcium pyrophosphate arthropathy associated chondrocalcinosis. There is chondrocalcinosis demonstrated within the
medial and lateral compartments of the left knee joint but relative preservation of joint spaces.
A.K. Brown / Best Practice & Research Clinical Rheumatology 27 (2013) 249–269
Fig. 13. Calcium pyrophosphate arthropathy. Chondrocalcinosis is present in the triangular fibrocartilage in the ulna-carpal regions
of both wrists, particularly the right. There is joint space loss at the right index finger MCP joint. Marked subchondral sclerosis, cysts,
joint space loss and osteophytes are present at both first CMC joints.
resembling a Charcot joint. Occasionally CPPD deposition may occur in soft tissue structures such as
tendons (calcific tendonitis) and bursae and be associated with the development of subcutaneous nodules.
Despite ongoing advances in imaging techniques and technologies, radiography continues to be a
key investigation in the initial assessment, diagnosis and ongoing management of patients with
arthropathy. It is generally safe, accessible and cost effective with the opportunity to provide timely and
useful information which is helpful to a range of health professionals. When considering the inflammatory arthropathies, one should keep in mind that radiographs do not provide specific information on
synovial or other soft tissue inflammation but can be useful in demonstrating characteristic pathological changes in bone and cartilage. As such their diagnostic utility may be greatest in established
rather than early disease. They remain the usual first-line imaging modality in providing information
on disease extent, severity and progression as part of ongoing disease surveillance.
*[1] Ionising Radiation (Medical Exposure). IR(ME)R 2000. The Department of Health.; 2000.
*[2] Goronzy JJ, Matteson EL, Fulbright JW, Warrington KJ, Chang-Miller A, Hunder GG, et al. Prognostic markers of radiographic progression in early rheumatoid arthritis. Arthritis & Rheumatism 2004;50(1):43–54.
[3] Scott DL, Pugner K, Kaarela K, Doyle DV, Woolf A, Holmes J, et al. The links between joint damage and disability in
rheumatoid arthritis. Rheumatology 2000;39(2):122–32.
[4] Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries JF, Cooper NS, et al. The American-Rheumatism-association 1987
revised criteria for the classification of rheumatoid-arthritis. Arthritis & Rheumatism 1988;31(3):315–24.
[5] Green M, Marzo-Ortega H, McGonagle D, Wakefield R, Proudman S, Conaghan P, et al. Persistence of mild, early inflammatory arthritis – the importance of disease duration, rheumatoid factor, and the shared epitope. Arthritis &
Rheumatism 1999;42(10):2184–8.
*[6] Aletaha D, Neogi T, Silman A, Funovits J, Felson DT, Bingham 3rd CO, et al. Rheumatoid arthritis classification criteria.
Arthritis & Rheumatism 2010;62(9):2569–81.
*[7] Sommer OJ, Kladosek A, Weiler V, Czembirek H, Boeck M, Stiskal M. Rheumatoid arthritis: a practical guide to state-ofthe-art imaging, image interpretation, and clinical implications. Radiographics 2005;25(2):381–98.
A.K. Brown / Best Practice & Research Clinical Rheumatology 27 (2013) 249–269
[8] NHS National Institute for Health and Clinical Excellence (NICE). Clinical guidelines CG79 rheumatoid arthritis: the
management of rheumatoid arthritis in adult.;
February 2009 [modified April 2009].
[9] Plant MJ, Jones PW, Saklatvala J, Ollier WER, Dawes PT. Patterns of radiological progression in early rheumatoid arthritis:
results of an 8 year prospective study. Journal of Rheumatology 1998;25(3):417–26.
[10] Drossaers-Bakker KW, Kroon HM, Zwinderman AH, Breedveld FC, Hazes JM. Radiographic damage of large joints in longterm rheumatoid arthritis and its relation to function. Rheumatology 2000;39:998–1003.
[11] Paulus HE, van der Heijde DMFM, Bulpitt KJ, Gold RH. Monitoring radiographic changes in early rheumatoid arthritis.
Journal of Rheumatology 1996;23(5):801–5.
[12] Machold KP, Stamm TA, Eberl GJM, Nell VKP, Dunky A, Uffmann M, et al. Very recent onset arthritis – clinical, laboratory,
and radiological findings during the first year of disease. Journal of Rheumatology 2002;29(11):2278–87.
[13] Lindqvist E, Jonsson K, Saxne T, Eberhardt K. Course of radiographic damage over 10 years in a cohort with early
rheumatoid arthritis. Annals of the Rheumatic Diseases 2003;62:611–6.
[14] Wakefield RJ, Gibbon WW, Conaghan PG, O’Connor P, McGonagle D, Pease C, et al. The value of sonography in the
detection of bone erosions in patients with rheumatoid arthritis – a comparison with conventional radiography. Arthritis
& Rheumatism 2000;43(12):2762–70.
[15] Backhaus M, Kamradt T, Sandrock D, Loreck D, Fritz J, Raber H, et al. Arthritis of the finger joints – a comprehensive
approach comparing conventional radiography, scintigraphy, ultrasound, and contrast-enhanced magnetic resonance
imaging. Arthritis & Rheumatism 1999;42(6):1232–45.
[16] Døhn UM, Ejbjerg BJ, Hasselquist M, Narvestad E, Court-Payen M, Szkudlarek M, et al. Rheumatoid arthritis bone erosion
volumes on CT and MRI: reliability and correlations with erosion scores on CT, MRI and radiography. Annals of the
Rheumatic Diseases 2007;66:1388–92.
[17] Hulsmans HMJ, Jacobs JWG, van der Heijde DM, van Albada-Kuipers GA, Schenk Y, Bijlsma JW. The course of radiographic
damage during the first six years of rheumatoid arthritis. Arthritis & Rheumatism 2000;43:1927–40.
*[18] Combe B, Landewe R, Lukas C, Bolosiu HD, Breedveld F, Dougados M, et al. EULAR recommendations for the management
of early arthritis: report of a task force or the European standing committee for international clinical studies including
therapeutics (ESCISIT). Annals of the Rheumatic Diseases 2007;66:34–45.
*[19] Roche CJ, Eyes BE, Whitehouse GH. The rheumatoid cervical spine: signs of instability on plain cervical radiographs.
Clinical Radiology 2002;57:241–9.
[20] Boini S, Guillemin F. Radiographic scoring methods as outcome measures in rheumatoid arthritis: properties and advantages. Annals of the Rheumatic Diseases 2001;60(9):817–26.
[21] Jansen LM, van der Horst-Bruinsma IE, van Schaardenburg D, Bezemer PD, Dijkmans BA. Predictors of radiographic joint
damage in patients with early rheumatoid arthritis. Annals of the Rheumatic Diseases 2001;60(10):924–7.
[22] Battistone MJ, Manaster BJ, Reda DJ, Clegg DO. The prevalence of sacroiliitis in psoriatic arthritis: new perspectives from a
large, multicenter cohort – a Department of Veterans Affairs Cooperative Study. Skeletal Radiology 1999;28(4):196–201.
*[23] McGonagle D, Conaghan PG, Emery P. Psoriatic arthritis: a unified concept twenty years on. Arthritis & Rheumatism 1999;
[24] Helliwell PS, Hetthen J, Sokoll K, Green M, Marchesoni A, Lubrano E, et al. Joint symmetry in early and late rheumatoid
and psoriatic arthritis – comparison with a mathematical model. Arthritis & Rheumatism 2000;43(4):865–71.
[25] Rahman P, Nguyen E, Cheung C, Schentag CT, Gladman DD. Comparison of radiologic severity in psoriatic arthritis and
rheumatoid arthritis. Journal of Rheumatology 2001;28:1041–4.
[26] Kane D, Stafford L, Bresnihan B, FitzGerald O. A prospective, clinical and radiological study of early psoriatic arthritis: an
early synovitis clinic experience. Rheumatology 2003;42(12):1460–8.
*[27] Rudwaleit M, van der Heijde D, Landewe. The development of Assessment of Spondyloarthritis international Society
classification criteria for axial spondyloarthritis (part II): validation and final selection. Annals of the Rheumatic Diseases
[28] Blum U, BuitragoTellez C, Mundinger A, Krause T, Laubenberger J, Vaith P, et al. Magnetic resonance imaging (MRI) for
detection of active sacroiliitis – a prospective study comparing conventional radiography, scintigraphy, and contrast
enhanced MRI. Journal of Rheumatology 1996;23(12):2107–15.
[29] Oostveen J, Prevo R, den Boer J, van de Laar M. Early detection of sacroiliitis on magnetic resonance imaging and subsequent development of sacroiliitis on plain radiography. A prospective, longitudinal study. Journal of Rheumatology
[30] van Tubergen A, Heuft-Dorenbosch L, Schulpen G, Landewe R, Wijers R, van der Heijde D, et al. Radiographic assessment
of sacroiliitis by radiologists and rheumatologists: does training improve quality? Annals of the Rheumatic Diseases
[31] Van der Linden S, Valkenburg HA, Cats A. Evaluation of diagnostic criteria for ankylosing spondylitis: a proposal for
modification of the New York criteria. Arthritis & Rheumatism 1984;27:361–8.
[32] Van der Heijde D, Sharp J, Wassenberg S, Gladman DD. Psoriatic arthritis imaging: a review of scoring methods. Annals of
the Rheumatic Diseases 2005;64(suppl. 2):ii61–4.
[33] Mau W, Zeidler H, Mau R, Majewski A, Freyschmidt J, Stangel W, et al. Clinical features and prognosis of patients with
possible ankylosing spondylitis. Results of a 10-year follow-up. Journal of Rheumatology 1988;15:1109–14.
*[34] Sieper J, Rudwaleit M, Baraliakos X, Brandt J, Braun J, Burgos-Vargas R, et al. The Assessment of Spondyloarthritis international Society (ASAS) handbook: a guide to assess spondyloarthritis. Annals of the Rheumatic Diseases 2009;
68(Suppl. II):ii1–44.
[35] Wanders AJ, Landewé RB, Spoorenberg A, Dougados M, van der Linden S, Mielants H, et al. What is the most appropriate
radiologic scoring method for ankylosing spondylitis? A comparison of the available methods based on the outcome
measures in rheumatology clinical trials filter. Arthritis & Rheumatism 2004;50:2622–32.
[36] Kellgran JH, Lawrence JS. Radiological assessment of osteo-arthritis. Annals of the Rheumatic Diseases 1957;16(4):494–
A.K. Brown / Best Practice & Research Clinical Rheumatology 27 (2013) 249–269
*[37] Conaghan PG, Hunter DJ, Maillefert JF, Reichmann WM, Losina E. Summary and recommendations of the OARSI FDA
osteoarthritis assessment of Structural change working group. Osteoarthritis Cartilage 2011;19(5):606–10. http://dx.doi.
org/10.1016/j.joca.2011.02.018. [Epub 2011 Mar 23].
AADI: anterior atlanto-dental interval
ACR: American College of Rheumatology
AP: antero-posterior
AS: ankylosing spondylitis
ASAS: Assessment of SpondyloArthritis international Society
BASRI: Bath Ankylosing Spondylitis Radiological Index
CMC: carpometacarpal
CPPD: calcium pyrophosphate crystal deposition
CT: computed tomography
EULAR: European League Against Rheumatism
NICE: National Institute of Clinical Excellence
MCP: metacarpophalangeal
MTP: metatarsophalangeal
MRI: magnetic resonance imaging
mSASSS: Modified Stoke Ankylosing Spondylitis Spinal Score
OA: osteoarthritis
OMERACT: Outcome Measures In Rheumatoid Arthritis Clinical Trials
PET: positron emission tomography
PIP: proximal interphalangeal
PsA: psoriatic arthritis
RA: rheumatoid arthritis
SASSS: Stoke Ankylosing Spondylitis Spinal Score
SPECT: single photon emission computed tomography
STIR: short tau inversion recovery
US: ultrasonography