The Pediatric Hip and Other Musculoskeletal Ultrasound Applications

History

The first in-depth use of sonography was performed by Graf, an Austrian orthopedic surgeon. Scanning was performed from the lateral approach with the femur in its anatomic position. Graf's method, which keyed on acetabular morphology, established ultrasound's ability to distinguish the cartilage, bone, and soft tissue structures that compose the immature hip joint. When real-time sonographic equipment became available, sonographers experimented with different approaches to the hip and the use of dynamic assessment. Both approaches recognize the need for critical landmarks of the femur and acetabulum. The dynamic technique, as proposed by Harcke and colleagues, mirrored the physical examination.

Although two basic assessment strategies evolved, morphologic and dynamic, the two methods have common features, in addition to stressing positional relationships and stability and assessment of critical acetabular landmarks. This formed the basis for an examination drawing on elements from both techniques, which has become part of the practice parameter recommended by the American College of Radiology (ACR), the American Institute of Ultrasound in Medicine (AIUM), Society of Pediatric Radiology (SPR), and Society of Radiologists in Ultrasound (SRU).

Technical Factors

Hip sonography should be performed with real-time linear array transducers. One should use the highest frequency transducer that provides adequate penetration of the soft tissues to the depth required. For infants up to 6 months of age, the 15-8 MHz broadband digital transducer is successful. A lower-frequency transducer may be required for infants older than 6 months.

Standard scanning is performed from the lateral or posterolateral aspect of the hip, moving the hip from the neutral position at rest to a position in which the hip is flexed. With the hip flexed, the femur is moved through a range of abduction and adduction, with stress views performed in the flexed position. One aspect of hip sonography relevant to dynamic examinations is the shift of the transducer between the examiner's hands when examining the right and left hip. The infant is lying supine with the feet toward the sonographer. When examining the left hip, the sonographer grasps the infant's left leg with the left hand, and the transducer is held in the right hand. When the right hip is examined, we recommend that the sonographer hold the transducer in the left hand and use the right hand to manipulate the infant's right leg. Although some sonographers find this awkward at first, ambidexterity in this context is easily mastered. We found that this technique makes it possible to perform the stress maneuver more reliably and better maintain the planes of interest.

For a satisfactory examination, the infant should be relaxed. Infants can be fed before or during the examination. Toys and other devices to attract the infant's attention are helpful and can be used as sonography is being performed. A parent can hold the infant's arms or head and can talk to the infant. There is no need for sedation. The upper body may remain clothed. Our standard practice is to leave the infant diapered and expose only the side of the hip being examined (strongly recommended for boys).

The anatomy is considered in orthogonal views. It is our routine practice to record images in each of these views for permanent records. This standardizes the examination and, in our institution, provides a guideline for the technologist who performs the initial examination. In describing views, we use a two-word combination that indicates the plane of the transducer with respect to the body (coronal or transverse) and the position of the hips (neutral or flexed).

It is the objective of the dynamic hip assessment to determine the position and stability of the femoral head, as well as the development of the acetabulum. With a normally positioned hip, the femoral head is congruently positioned within the acetabulum. Mild displacement, such as when the head is in contact with part of the acetabulum or is displaced but partly covered, is referred to as subluxation. The dislocated hip has no contact with or coverage by the acetabulum. A change in position of the femur may change the relationship of the femoral head and acetabulum. A hip that is subluxated in the neutral or rest position may seat itself with flexion and abduction. A dislocated hip may improve its position and partially reduce to a subluxated position. This is, in fact, a principle of treatment.

The stability of the hip is determined through motion and the application of stress. The stress maneuvers are the imaging counterparts of the clinical Barlow and Ortolani maneuvers, which are the basis for the clinical detection of a hip abnormality. The Barlow test determines whether the hip can be dislocated. The hip is flexed and the thigh brought into the adducted position. The gentle push posteriorly can demonstrate instability by causing the femoral head to move out of the acetabulum. The Ortolani test determines whether the dislocated hip can be reduced. As the flexed, dislocated hip is abducted into a frog-leg position, the examiner feels a vibration or “clunk” that results when the femoral head returns to the acetabulum. The normal hip is always seated at rest, with motion, and during the application of stress. The lax hip is normally positioned at rest and shows mild subluxation with stress. It must, however, invariably remain within the confines of the acetabulum. The subluxated hip is displaced laterally at rest and is loose; it may be dislocatable with stress and reducible with flexion and abduction. The dislocated hip may be able to be returned to the acetabulum with traction and abduction. This hip is distinguished from the most severe form of DDH, in which the femoral head is dislocated and cannot be reduced.

At birth, the proximal femur and much of the acetabulum are composed of cartilage. On sonographic examination, cartilage is hypoechoic compared with soft tissue, so it is easy to distinguish. A few scattered specular echoes can be visualized within the cartilage when high-frequency transducers are used and technique adjustments are optimally set. The acetabulum is composed of both bone and cartilage. At birth, the bony ossification centers in the ilium, ischium, and pubis are separated by the triradiate cartilage, which has a Y configuration. A cartilaginous acetabular rim (the labrum ) extends outward from the acetabulum to form the cup that normally contains the femoral head. Most of the acetabular cartilage has an echogenicity similar to the femoral head. It is still possible to determine the joint line, which distinguishes the cartilaginous acetabulum from the femoral head, by simply rotating the femur. More pronounced movement of the hip often creates echoes within the joint space, probably as a result of the formation of microbubbles. At the lateral margin of the labrum, the hyaline cartilage changes to fibrocartilage, and this shows increased echogenicity. The echogenic hip capsule, composed of fibrous tissue, borders the femoral head laterally. The bony components of the hip reflect all of the sound beam from their surface. This creates a bright linear or curvilinear appearance on the sonogram, indicating the contour of the osseous surfaces in that plane.

Radiographically, the ossification center of the femoral head is recognized between the second and eighth months of life. It is typically seen earlier in females than in males, and there is a wide normal variation for the time of appearance. Although some asymmetry between the left and right hips can be normal, both in time of appearance and in size, delayed appearance and development are associated with DDH. Hip sonograms reflect the development of the ossification center and can be used to document the development of the center. The ossification center can be found with ultrasound several weeks before it is visible radiographically. Initially, a confluence of blood vessels produces increased echoes within the cartilage. This precedes actual ossification. As ossification begins, the calcium content is insufficient to produce a visible radiographic density; however, the sound waves are reflected. With maturation, the size of the ossification center increases. In early development, the echoes from the center have a punctate appearance, whereas later in the first year of life, the growth in size gives it a curvilinear margin. As the normal infant approaches 1 year of age, the size of the ossification center precludes accurate determination of medial acetabular landmarks.

Sonography of the hip is practical only up to 8 months of age, unless there is delayed ossification of the femoral head. Between 4 months and 6 months, radiography becomes more reliable. Usually by 1 year of age, the femoral ossification center is large enough to prevent good sonographic visualization of the acetabulum. The presence and size of the ossific nucleus can be evaluated in all views.

Coronal/Neutral View

The coronal/neutral view, which forms the basis for the morphologic technique, is performed from the lateral aspect of the joint with the plane of the ultrasound beam oriented coronally with respect to the hip joint. The femur is maintained with a physiologic amount of flexion. The coronal/neutral view is performed with the patient supine. The transducer is placed on the lateral aspect of the hip, and the hip is scanned until a standard plane of section is obtained ( Fig. 55.1 ). The plane must precisely demonstrate the midportion of the acetabulum, with the straight iliac line superiorly and the inferior tip of the os ilium seen medially within the acetabulum. The echogenic tip of the labrum should also be visualized. The alpha and beta angles, if measured, relate to fixed points on the bony and cartilaginous components of the acetabulum (see Fig. 55.1 ), and the exact plane must be obtained for the measurements to be reliable. The similarity can be noted between the appearance of the acetabulum in this coronal/neutral view and in the coronal/flexion view ( Fig. 55.2 ). The difference is that the bony shaft (metaphysis) of the femoral neck is visualized below the femoral head in the coronal/neutral projection. In the coronal/flexion view, the femoral shaft is not in the plane of examination because the femur is flexed. A stability test can be performed in the coronal/neutral view by gently pushing and pulling the infant's leg. This helps to verify deformity of the acetabulum and to identify craniodorsal movement of the femoral head under pressure. Zieger and colleagues proposed a further adaptation of this view, advocating flexion and adduction of the hip to identify lateral displacement when instability is present. This is similar to the coronal/flexion stress view.

In the normal coronal/neutral view, the femoral head is resting against the bony acetabulum. The acetabular roof should have a concave configuration and cover at least half the femoral head. The cartilage of the acetabular roof is hypoechoic and extends lateral to the acetabular lip, terminating in the labrum, which is composed of fibrocartilage and is echogenic (see Fig. 55.1C ). When a hip becomes subluxated or dislocated, the femoral head gradually migrates laterally and superiorly, with progressively decreased coverage of the femoral head (see Fig. 55.1E ). In hip dysplasia, the acetabular roof is irregular and angled, and the labrum is deflected superiorly and becomes echogenic and thickened. When the hip is frankly dislocated, the labrum may be deformed. Echogenic soft tissue is interposed between the femoral head and the bony acetabulum. A combination of deformed labrum, fibrofatty tissue (pulvinar), and thickened ligamentum teres prevents the hip from being reduced.

The acetabulum can be assessed visually or with measurements, noting the depth and angulation of the acetabular roof, as well as the appearance of the labrum (see Fig. 55.1F ). This can be seen in both coronal/neutral and coronal/flexion views and should be described in the report. Morin and colleagues correlated coverage of the femoral head by the bony acetabulum with measurements of the acetabular angle. This assessment relates acetabular depth (d) to the diameter of the femoral head (D) and is expressed as percent (d/D × 100) coverage of the femoral head. These data showed that normal radiographic measurements always had a femoral head coverage that exceeded 56%, and that clearly abnormal radiographic measurements had coverage of less than 40%. This information should be used with caution because there are intermediate values for which sonographic and radiographic measurements do not correlate. We have also noted cases in which sonography showed the acetabulum to be better developed than it appeared radiographically, and vice versa. This discrepancy occurs because the radiograph is a two-dimensional projection of the three-dimensional pelvis, and selected sonographic images illustrate a single coronal slice that may not match the projection.

Classification of hip joints may also be based on the measurement of the alpha and beta angles (see Fig. 55.1D-F ). The alpha angle measures the inclination of the osseous acetabular roof with respect to the lateral margin of the iliac bone (baseline). The beta angle is formed by the baseline iliac bone and the inclination of the anterior cartilaginous acetabular roof, for which the tip of the labrum is its key landmark. Ultrasound units may contain software that facilitates measurement of these angles. Four basic hip types have been proposed on the basis of alpha and beta measurements. Most of these subtypes have been subdivided, and small changes in angular measurements can result in a change in category. The reproducibility of angular measurements and subtypes has been a point of considerable discussion. In Europe, classification by measurement has been based on large numbers of infants examined. Some examiners have experienced problems with the use of measurements, but those who adhere strictly to the technique find acceptable reproducibility.

Coronal/Flexion View

In the coronal/flexion view the transducer is maintained in a coronal plane with respect to the acetabulum ( Fig. 55.2A ) while the hip is moved to a 90-degree angle of flexion. During the assessment in this view, the transducer is moved in an anteroposterior direction with respect to the body to visualize the entire hip. Anterior to the femoral head, the curvilinear margin of the bony femoral shaft is identified. In the midportion of the acetabulum, the normally positioned femoral head is surrounded by echoes from the bony acetabular components (see Fig. 55.2B ). Superiorly, the lateral margin of the iliac bone is seen, and the transducer position must be adjusted so the iliac bone becomes a straight horizontal line on the monitor. This landmark ( iliac bone line: flat and straight) is a key to accurately visualizing the midacetabulum and to obtaining the maximum depth of the acetabulum. When the transducer is positioned too anteriorly, the iliac line is inclined laterally; and if positioned too posteriorly, the iliac line exhibits concavity. When the plane is not correctly selected, it could be falsely concluded that the acetabulum is maldeveloped. A normal hip gives the appearance of a “ball on a spoon” in the midacetabulum. The femoral head represents the ball, the acetabulum forms the bowl of the spoon, and the iliac line is the handle. When the transducer is moved posteriorly and the scan plane is over the posterior margin of the acetabulum, the posterior limb of the triradiate cartilage, also known as the “posterior lip,” becomes an easily recognized landmark. The bone above and below the cartilage notch is flat, and the normally positioned femoral head is not visualized ( Videos 55.1 and 55.2 ).

In subluxation, the femoral head is displaced laterally, posteriorly, or both, with respect to the acetabulum. Soft tissue echoes are seen between the femoral head and the bony reflections from the medial acetabulum. In dislocation, the femoral head is completely out of the acetabulum (see Fig. 55.2C ). With superior dislocations, the femoral head may rest against the iliac bone. In posterior dislocations, the femoral head is seen lateral to the posterior lip. The acetabulum is usually not visualized in a dislocation because the bony shaft of the femur blocks the view.

Dynamic examination in the coronal/flexion view has two components. The first is performed over the posterior lip using a push-and-pull maneuver (see Fig. 55.2D-E ). In the normal hip the femoral head is never seen over the posterior lip of the acetabulum. When there is instability, a portion of the femoral head appears over the posterior lip of the triradiate cartilage as the femur is pushed. With a pull, the head disappears from the plane. In a dislocated hip, the femoral head may be located over the posterior lip and may or may not move out of the plane with traction. The second component of the dynamic examination is performed over the midacetabulum. The Barlow-type maneuver is performed with adduction and gentle pushing against the knee. In the normal hip the femoral head will remain in place against the acetabulum. With subluxation or dislocation, the head will migrate laterally and posteriorly, and there will be echogenic soft tissue between the femoral head and the acetabulum ( Videos 55.3 , 55.4 , and 55.5 ).

Transverse/Flexion View

Transition from the coronal/flexion view to the transverse/flexion view is accomplished by rotating the transducer 90 degrees and moving the transducer posteriorly so it is in a posterolateral position over the hip joint. The orientation of the scan plane is parallel to the flexed femoral shaft ( Fig. 55.3A ). The plane of the transducer and the landmarks are demonstrated on a computed tomography (CT) scan of a patient in a spica cast with a normal left and dislocated right hip (see Fig. 55.3B ). Sonographically, the bony shaft and metaphysis of the femur give bright reflected echoes anteriorly, adjacent to the sonolucent femoral head. The echoes from the bony acetabulum appear posterior to the femoral head, and in the normal hip, a -shaped configuration is produced by the metaphysis and the ischium (see Fig. 55.3C ). The relationship of the femoral head and acetabulum is observed while the flexed hip is moved from maximum adduction to wide abduction. The sonogram changes its appearance in abduction and adduction. The deep, U -shaped configuration is produced with maximum abduction, whereas in adduction, a shallower, V -shaped appearance is observed. It is important to have the transducer positioned posterolaterally over the hip to see the medial acetabulum. When the transducer is not posterior enough, the view of the acetabulum is blocked by the femoral metaphysis, and the hip can appear falsely displaced. In adduction, the hip is stressed with a gentle posterior push (a Barlow test ). In the normal hip, the femoral head will remain deeply in the acetabulum in contact with the ischium with stress. In subluxation, the hip will be normally positioned or mildly displaced at rest, and there will be increased lateral displacement from the medial acetabulum with stress, but the femoral head will remain in contact with a portion of the ischium. With frank dislocation, the hip will be laterally and posteriorly displaced to the extent that the femoral head has no contact with the acetabulum, and the normal U -shaped configuration cannot be obtained (see Fig. 55.3D ). The process of dislocation and reduction is able to be visualized in unstable hips in the transverse/flexion view. With abduction, the dislocated hip may be reduced, representing the sonographic counterpart of the Ortolani maneuver. With adduction, the subluxated hip may be dislocatable, corresponding to the Barlow maneuver ( Videos 55.6 , 55.7 , and 55.8 ).

Alternative Views

A transverse/neutral view is still used in our examination protocol, but this view has not been included in published guidelines. The transducer is directed horizontally into the acetabulum from the lateral aspect of the hip; the leg is in physiologic neutral position. The plane of interest is one that passes through the femoral head into the acetabulum at the center of the triradiate cartilage. The position of the head with respect to the triradiate cartilage is evident and can confirm findings on the coronal/neutral view.

A number of anterior views have been described, and sonographers experienced in their use have indicated success with these views. The anterior approach of Dahlstrom and colleagues is performed with the patient supine and the hips flexed and abducted. The transducer is placed anterior to the hip joint and is centered over the femoral head with the plane of the sound beam parallel to the femoral neck ( Fig. 55.4A-B ). A Barlow or push maneuver can be performed to detect instability. Complete dislocation is considered to be present when femoral head displacement exceeds 50% of its diameter (see Fig. 55.4C ). The image produced in a normal hip is an axial section through the acetabulum and a longitudinal section through the femoral head and neck. The anterior view is particularly useful in rigid abduction splints and casts in which the posterior aspect of the hip is covered.