Perthes disease

Radiographic features

The best initial test for the diagnosis of Perthes is a pelvic radiograph. In a small number of patients
with Perthes, the radiograph will be normal and persistent symptoms will trigger further imaging,
e.g. MRI.

The investigation of atraumatic limp will often include a hip ultrasound to look for effusion, but
ultrasound is unlikely to pick up osteonecrosis.

The radiographic findings are those of osteonecrosis. There are separate systems for staging of
Perthes disease:

temporal evolution

Waldenström classification

modified Elizabethtown classification

severity and prognosis

Catterall classification: extent of femoral head involvement

Salter-Thompson classification: extent of femoral head involvement

Herring classification: lateral pillar involvement

healed stage deformity: osteoarthritis risk

Stulberg classification

Plain radiograph

The radiographic changes to the femoral epiphyses depend on the severity of osteonecrosis and the
amount of time that there has been an alteration of blood supply:

early: there may be no appreciable change

established: reduction in epiphysis size, lucency

late: fragmentation, destruction

As changes progress, the width of the femoral neck increases (coxa magna) in order to increase
weight-bearing support.

Early signs

joint effusion: widening of the medial joint space

asymmetrical femoral epiphyseal size (smaller on the affected side)

apparent increased density of the femoral head epiphysis

blurring of the physeal plate (stage 1)

radiolucency of the proximal metaphysis

Late signs

Eventually, the femoral head begins to fragment (stage 2), with subchondral lucency (crescent sign)
and redistribution of weight-bearing stresses leading to thickening of some trabeculae which
become more prominent.

The typical findings of advanced burnt out (stage 4) Perthes disease are:

femoral head deformity with widening and flattening (coxa plana)

proximal femoral neck deformity: coxa magna

"sagging rope sign" (thin sclerotic line running across the femoral neck)

Additionally, tongues of cartilage sometimes extend inferolaterally into the femoral neck, creating
lucencies, which must be distinguished from infection or neoplastic lesions 4. The presence of
metaphyseal involvement not only increases the likelihood of femoral neck deformity but also makes
early physeal closure with resulting leg length disparity more likely.

Arthrography

Traditionally arthrography performed under general anesthesia with conventional fluoroscopy is

performed to assess congruence between the femoral head and the acetabulum in a variety of
positions 3. MRI is increasingly replacing this, in an effort to eliminate pelvic irradiation.

DDH

Radiographic features

For imaging assessment of developmental dysplasia of the hip, ultrasound is the modality of choice
prior to the ossification of the proximal femoral epiphysis. Once there is a significant ossification
then an x-ray examination is required.

For some reason, the left hip is said to be more frequently affected 4. One-third of cases are affected
bilaterally 5.

Ultrasound

Ultrasound is the test of choice in the infant (<6 months) as the proximal femoral epiphysis has not
yet significantly ossified. Additionally, it has the advantage of being a real-time dynamic examination
allowing the stability of the hip to be assessed with stress views.

Some values are used to 'objectively' assess morphology.

Alpha angle

The alpha angle is formed by the acetabular roof to the vertical cortex of the ilium. This is a similar
measurement to that of the acetabular angle (see below). The normal value is greater than or equal
to 60º.

Beta angle

The beta angle is formed by the vertical cortex of the ilium and the triangular labral fibrocartilage
(echogenic triangle). The normal value is less than 77º 6 but is only useful in assessing immature hips
when combined with the alpha angle.

Bony coverage (d:D ratio)

The percentage of the femoral epiphysis covered by the acetabular roof. A value of >50% is
considered normal 7, 14.

Plain radiograph

A single AP radiograph is the most appropriate examination in children where femoral head
ossification has occurred, e.g. over 1 year old. A frog-leg lateral view does not add additional
information but does double the radiation dose 15.

Look for symmetry of the femoral head ossification center (delayed on the abnormal side).
Determine the relationship of the proximal femur to the developing pelvis. The femoral head should
be centered in the inferomedial quadrant defined by the intersection of Hilgenreiner line and Perkin
line.

Interruption of the Shenton line may also be evident.

The acetabular angle should be ~30° at birth and progressively reduce with the maturation of the
joint.

The extrusion index is a percentage measure of bony coverage of the femoral head by acetabulum in
patients with fully matured femoral epiphyses. A value of <25% has been reported as normal 11. The
greater the degree of acetabular dysplasia, the greater the extrusion index.

The center-edge angle (CEA) of Wiberg may be used in younger children. An angle is formed by
Perkin line and a line from the center of the femoral head to the lateral edge of the acetabulum
where a value of < 20° is considered abnormal but considered only reliable in patients > 5 years of
age
AVN

Radiographic features

A specific staging system (Ficat staging) exists for the hip which includes x-ray, MRI and bone scan
appearances, and covers much of the imaging appearances, thus please refer to that article.

Other than describing the general appearance of the affected region, the following are necessary to
include in the report as they have a bearing on prognosis and treatment:

position

estimating percentage volume of the head involved (axial) and percentage weight-bearing surface
involved (coronal)

coexisting osteoarthritis or secondary degenerative change

joint effusion

presence of a potentially unstable osteochondral fragment: rim sign

subchondral fractures
CT

Often more sensitive than plain film in showing subchondral fractures.

MRI

MRI is the most sensitive modality, with a sensitivity of 71-100% and specificity of 94-100% 1. As
there is a high rate of bilateral involvement, both hips should be included in the field of view of at
least some sequences.

T1: usually the initial specific findings are areas of low signal intensity representing edema, which
can be bordered by a hyperintense line that represents blood products

T2: may show a second hyperintense inner line between normal marrow and ischemic marrow. This
appearance is highly specific for AVN hip and is known as "double-line sign".

The Mitchell classification is commonly used to classify AVN based on MR images

OA HIP

Classification

Plain radiograph

Different grading schemes are described for plain radiographs of the hip:

grade 0: normal

grade 1: possible joint space narrowing and subtle osteophytes

grade 2: definite joint space narrowing, defined osteophytes and some sclerosis, especially in the
acetabular region

grade 3: marked joint space narrowing, small osteophytes, some sclerosis and cyst formation and
deformity of femoral head and acetabulum

grade 4: gross loss of joint space with above features plus large osteophytes and increased deformity
of the femoral head and acetabulum

Tönnis classification for osteoarthritis of the hip

The original Tönnis classification consists of three degrees of degenerative changes featuring
osteoarthritis of the hip. Grade 0 was added later 2,3.

grade 0: no signs of osteoarthritis

grade 1: minor joint space narrowing, subchondral sclerosis of the femoral head and/or acetabulum,
small osteophytes

grade 2: moderate joint space narrowing, small subchondral cysts of the femoral head and/or
acetabulum, moderate loss of sphericity of the femoral head

grade 3: severe joint space narrowing or obliteration, large subchondral cysts, severe deformity of
the femoral head

Croft score for hip osteoarthritis

The Croft score uses different compositions of the following features for the grading of hip
osteoarthritis: osteophytes, joint space narrowing, subchondral sclerosis, subchondral cysts 4,5.

grade 1: only osteophytic changes

grade 2: only joint space narrowing

grade 3: two of the above mentioned radiographic features

grade 4: three of the above mentioned radiographic features

grade 5: femoral head deformity plus three of the above mentioned radiographic features

MRI
grade 0: normal

grade 1: inhomogeneous high signal intensity in cartilage (T2WI)

grade 2: inhomogeneity with areas of high signal intensity in articular cartilage (T2WI); indistinct
trabeculae or signal intensity loss in femoral head and neck (T1WI)

grade 3: criteria of grade 1 and 2 plus indistinct zone between femoral head and acetabulum;
subchondral signal loss due to bone sclerosis

grade 4: above criteria plus femoral head deformity

ANH

Radiographic features

The axial skeleton is predominantly affected, although in ~20% of cases the peripheral joints are also
involved.

Plain radiograph

Sacroiliac joints

sacroiliitis is usually the first manifestation 5 and is symmetrical and bilateral

the sacroiliac joints first widen before they narrow

subchondral erosions, sclerosis, and proliferation on the iliac side of the SI joints

at end-stage, the SI joint may be seen as a thin line or not visible

Spine

early spondylitis is characterized by small erosions at the corners of vertebral bodies with reactive
sclerosis: Romanus lesions of the spine (shiny corner sign)

vertebral body squaring

noninfectious spondylodiscitis: Andersson lesion

diffuse syndesmophytic ankylosis can give a "bamboo spine" appearance

syndesmophytes are classically described as paravertebral ossification running parallel to the spine

linear ossification along the central spine; representing interspinous ligament ossification can give a
"dagger spine" appearance on frontal radiographs;

ossification of spinal ligaments, joints and discs (with fatty marrow within the ossified disc, best seen
on MRI)

apophyseal and costovertebral arthritis and ankylosis

enthesophyte formation from enthesopathy

pseudoarthroses may form at fracture sites

dural ectasia

Hips

Hip involvement is generally bilateral and symmetric, with uniform joint space narrowing, axial
migration of the femoral head sometimes reaching a state of protrusio acetabuli, and a collar of
osteophytes at the femoral head-neck junction.
Pelvis

Whiskering of the pelvic bones primarily affects the ischial tuberosities, resulting from ossification of
the ligamentous origins.

There can be bridging or fusion of the pubic symphysis.

Knees

Knees demonstrate uniform joint space narrowing with bony proliferation.

Hands

Hands are generally involved asymmetrically, with smaller, shallower erosions and marginal
periostitis.

Shoulders

glenohumeral joint involvement is not uncommon and demonstrates a large erosion of the
anterolateral aspect of the humeral head, producing a 'hatchet' deformity ref

marrow edema of the acromion process, at the site of origin of deltoid muscle, has been described
as a very specific sign of the disease ref
Chest

Radiographs of the lungs may demonstrate progressive fibrosis and bullous changes at the apices.
These lesions may resemble tuberculosis infection and bullae may become infected.

Cardiac

Plain radiograph may be normal, or may reveal cardiomegaly.

CT

may be useful in selected patients with normal or equivocal findings on sacroiliac joint radiographs

chronic structural changes such as joint erosions, subchondral sclerosis, and bony ankylosis are
better visualized on CT than on MRI or radiographs 15-17

some normal variants of the SI joints may mimic features of sacroiliitis

supplements scintigraphy in evaluating areas of increased uptake

superior to radiographs and MRI in demonstrating injuries

imaging modality of choice in patients with advanced ankylosing spondylitis in whom there is
suspicion of cervical spine fracture

sagittal reformats should be obtained as axial images poorly assess the transverse fracture plane

MRI

may have a role in early diagnosis of sacroiliitis; MRI is more sensitive than CT or plain radiography in
detecting inflammatory changes (which precede structural changes) such as bone marrow edema
(best demonstrated on STIR sequences), synovitis and capsulitis (on gadolinium enhanced T1
weighted sequences) 16,18

synovial enhancement on MR correlates with disease activity measured by inflammatory mediators

enhancement of the interspinous ligaments is indicative of enthesitis

increased T2 signal correlates with edema or vascularized fibrous tissue

superior to CT in the detection of cartilage inflammation and destruction

useful in following treatment results in patients with active ankylosing spondylitis

Bone scintigraphy

maybe helpful in selected patients with normal or equivocal findings on sacroiliac joint radiographs

qualitative assessment of accumulation of radionuclide

e in the SI joints may be difficult due to normal uptake in this location; thus, quantitative analysis
may be more useful
ratios of SI joint to sacral uptake of 1.3:1 or higher is abnormal