Developmental dysplasia of the hip (DDH), which used to refer to the term “congenital hip dysplasia,” is not very accurate. The femoral head may be in normal position, subluxed or dislocated. Reports based on physical examination show that the incidence of this lesion varies widely worldwide, accounting for approximately 2 to 6 in 1000 surviving newborns.
The causes of hip dysplasia are multifactorial and appear to be related to late changes during pregnancy to what would normally be a normally formed hip joint. Various physical factors (female, Caucasian, etc.), mechanical factors (low amniotic fluid, breech delivery, etc.) and functional factors (maternal estrogen levels blocking collagen maturation, familial, etc.) combine to cause progressive displacement of the femoral head out of the acetabulum. From a pathophysiological point of view, the maintenance of a normal relationship between the acetabulum and the femoral head is a guarantee of normal development of the acetabulum.
In neonates, reducing prenatal dislocation facilitates their development into a stable hip joint. However, without early recognition of hip dislocation, certain adaptive changes can make it more difficult for the femoral head to retract and reset. In particular, the hip muscles remain tense and shortened for long periods of time because they cannot be at their normal resting length, the acetabulum loses its normal concave shape, the joint space is filled with fibrofatty tissue, and the round ligaments and joint capsule become long and lax. In other words, it is difficult to retract and reset the femoral head by relying on simple manual repositioning. Hua Xing, Ultrasound Department, Chongqing Southwest Hospital
To avoid the impact on the child, the family and even the health care system, early diagnosis of DDH is crucial to obtain early and relatively easy therapeutic measures for it. Physical examination is a key factor in identifying the disease: it includes visual examination (shortening of the limb due to superior femoral dislocation, excess skin folds, loss of slight hip and knee flexion when supine, etc.) and two basic stress tests: the Ortolani test, which detects retraction of the dislocated hip; and the Barlow test, which attempts to elicit a dislocation of the femoral head that is in a normal position.
However, the accuracy of the physical examination is not completely certain, with a reported misdiagnosis rate of less than 1%. Because of the shortcomings of X-ray in imaging early hip dysplasia, such as ionizing radioactivity and inability to image the associated unossified structures, ultrasound has thus become an alternative imaging technique and has now changed the understanding and treatment pathway for DDH. The two currently prevailing ultrasound methods reflect two philosophies: one based on morphologic criteria (Graf method) and the other based on dynamic examination (Harcke method.) In 1993, two authors consensually proposed a standard examination method that combines the various features of both techniques.
1. Graf technique
This technique is based on the examination technique elaborated by Graf in 1980 and currently in use in Europe. Hip ultrasonography is performed in the infant’s lateral recumbent position, which is the optimal position for both positioning and for fixation and support of the stress maneuver test. A number of positioning devices have also been designed to make the examination easier, although these devices are not required. The coronal view of the acetabulum at the deepest point of the acetabulum is the standard reference view used for measurement. This view is best obtained with the knee in mild flexion, although the hip can also be positioned in neutral or mild internal rotation as originally described by Graf. The ultrasound scan access route is located slightly posterior to the lateral aspect of the hip.
The relevant anatomic structures identified by ultrasound range from superficial to deep and include: the gluteus medius and gluteus minimus muscles, the joint capsule (covering the femoral head); the femoral head (which appears as a circular hypoechoic image with punctate echogenicity); the Y-shaped cartilage, or triangular cartilage (which appears as the lowest point of the acetabulum); the ilium (a linear hyperechoic extension above the acetabulum forming the iliac wing); and the iliac promontory (the point where the iliac wing meets the acetabular roof). On a correct coronal projection view (with the lowest point of the acetabular fossa as the axis of spin), the iliac promontory must be shown as clearly and sharply as possible, with the femoral head at its larger diameter and the iliac wing in a vertical plane parallel to the probe. If the sweep is too far forward, the iliac wing will face outward; if the sweep is too far back, the iliac wing will be concave (gluteal fossa). In general, the sharper the iliac promontory, the more mature the hip joint development.
On top of the standard coronal view, Graf proposed two angles formed by the intersection of three lines: the iliac line (baseline), the iliac wing tangent; the acetabular roof line, the line connecting the iliac promontory to the deepest edge of the acetabulum; and the glenoid lip line, the line connecting the iliac promontory to the center of the fibrocartilaginous glenoid lip. The intersection of the first two lines forms the alpha angle (acetabular inclination angle), which reflects the depth of the bony acetabular roof and the coverage of the femoral head.
This angle is important because it correlates with hip maturity; the smaller the angle, the greater the degree of dysplasia. In a normal mature hip, the alpha angle should be ≥60°. The second angle, the beta angle (cartilage apex angle), is derived from the intersection of the baseline and glenoid labrum lines. A normal infant beta angle should be <55°. The beta angle is indicative of the degree of upward displacement of the femoral head; the greater the angle, the less the cartilaginous apex covers the femoral head.
Graf classifies hip dysplasia into four main types (I to IV), a typology derived from a combined measurement of the above angles. It should be noted that these types represent a continuum of hip manifestations from normal to severe dysplasia and are not four distinct groups independent of each other. graf type I (α ≥ 60°) suggests a mature hip with a good skeletal configuration, a sharp iliac promontory, and a well-covered cartilaginous apex triangle. Graf type II (50° < α < 60°) suggests a satisfactory skeletal configuration with a rounded iliac promontory and a well-covered cartilaginous apex triangle. type II hips can be seen in both physiologically immature infants younger than 3 months (type IIA) and in infants with delayed ossification ( types IIB and IIC).
Type IIA can be subdivided into two subtypes, IIA(+) (physiologic, age-appropriate) and IIA(-) (maturation defect). The “off-center” hip has a severely deficient skeletal configuration, with a rounded or flat iliac promontory and a triangular displacement of the cartilage apex, and is designated as type IID. In immature hips (type IIA), 95% of IIA (+) and 84% of IIA (-) spontaneously resolve (mature). However, clinical observation and follow-up are still needed. Children with mild hip dysplasia (43°<α<49°) (types IIC and IID) usually require treatment, although some authors believe that such hips tend to develop to normal on their own if they are stable.
Graf type III (α<43°) and type IV (α<43° or unmeasured) hips are eccentric subluxed or dislocated hips with poor skeletal configuration, flat iliac promontory, and triangular displacement of the cartilaginous apex: these two types of hips require prompt treatment. Although the reproducibility and reliability of angular measurements remain controversial, most stable hips (types I and IIA) can be easily and accurately identified by the morphology of the acetabular structures alone, rather than based on quantitative assessment: this will reduce the time required for most imaging examinations. Apart from the angulation, the normal hip joint accommodates half of the femoral head within the acetabulum: therefore, the extension line of the iliac line should normally pass through the middle of the femoral head. Femoral head coverage decreases with subluxation. Thickening of the acetabular cartilage (>3.5 mm) has been reported in patients with DDH.
2. Harcke’s technique
Harcke first proposed dynamic ultrasonography of the infant hip. Compared with Graf’s method, Harcke’s technique, which is widely used in the United States, focuses on the instability of the hip joint as the main abnormal manifestation rather than morphological manifestation. The child is placed supine and a four-step examination technique is used: coronal and cross-sectional images of the hip in neutral and flexion positions are obtained under resting conditions and during a stress test, respectively. In the initial elaboration, the coronal-neutral view images duplicated Graf’s standard planes, but without angular measurements.
In the coronal-flexion view, the probe is placed slightly posterior to the deltoid ligament in the standard coronal view. Performing the stress test maneuver: The femoral head is pistoned by a “push” and “pull” maneuver in knee flexion, which simulates the Barlow test. In this condition, the head of the femur does not appear above the posterior lip of the acetabulum in normal conditions; in subluxation, the head of the femur partially rises above the posterior lip of the acetabulum during the application of pressure. The transverse section-flexion plane confirmed that the femoral head was located between the bony situs and the medial part of the acetabulum.
In the unstable hip, incomplete dislocation of the femoral head behind the ilium was seen when the inversion (Barlow’s test maneuver) or abduction (Ortolani’s maneuver) was performed with gentle posterior pushing, whereas no displacement of the normal hip joint occurred. The transverse-neutral view shows that the center of the acetabulum is located at the level of the Y-shaped cartilage. The femoral head is normally located in the acetabulum. In case of dislocation, the Y-shaped cartilage cannot be observed in this plane. During the first 2 weeks of life, laxity within the normal range may result in a mild posterior displacement of the femur.
Using the Harcke technique, the hip can be classified as normal, relaxed under pressure, subluxed, or dislocated. Although both the Graf and Harcke methods have shown comparable results in practice, it appears that the dynamic examination technique requires more training and practice for the examiner compared to the Graf technique.
3. Femoral head coverage technique
A third method to assess the degree of femoral head outgrowth was proposed by Morin et al. and then revised by Terjesen et al. Based on two lines drawn on a coronal section (the same as the baseline in Graf’s method), one (d) representing the distance from the baseline to the medial surface of the femoral head and the other (D) representing the maximum diameter of the femoral head, the percentage of femoral head coverage of the bony acetabulum (“bone margin percentage” or “femoral head coverage”) can be obtained by the equation (d/D) × 100%. A femoral head coverage of less than 50% is considered abnormal. This method cannot be used for dislocated hips because the normal relationship between the acetabulum and the femoral head is lost in this case.
4. Screening procedures and follow-up
Establishing screening procedures for DDH to avoid delayed diagnosis and missed diagnoses is a complex task that is controversial and has not yet been fully resolved. In general, the ideal screening strategy should allow for early detection of all affected cases at a reasonable cost and without false positives. Screening by clinical examination alone has been shown to reduce late morbidity by 50%, but does not eliminate false positives (leading to overtreatment) and false negatives (leading to late onset). Ultrasound has been chosen as a screening tool in many countries within the last 10-15 years. The technique has been shown to detect one third more abnormal cases than clinical examination, and newborns with normal hips on ultrasound have a lower chance of developing DDH in the future. However, screening methods are not standardized and may be inconsistent from country to country and from center to center in the same country. Some European countries perform universal screening, while the United States selectively screens newborns with known risk factors.
Universal screening using the Graf method shows that 75-85% of newborns have normal hips, 13-25% have immature hips, and 2-4% have dysplastic hips. As for the relationship between morphological presentation and stability, only 0.1% of morphologically normal hips may develop dislocation, whereas 0.6% of immature hips, 64% of mildly dysplastic and almost 100% of severely dysplastic hips may develop dislocation. In Italy, a remarkable absence of false negatives has been obtained since the introduction of universal screening in 1987.
On the other hand, selective screening limited to infants with risk factors or physical findings of unstable hips does not eliminate delayed onset in infants with normal physical examination at birth and without risk factors (about 0.025-0.035%). As for the optimal timing of screening, it is currently believed that screening at 4-6 weeks after birth detects many immature or unstable hips that recover on their own. This is why some authors argue for delaying ultrasound in unselected infants. However, delaying screening may also lead to delayed treatment and may also result in a higher than expected percentage of hips that will be treated.
In general, the economic analysis of the screening process is more difficult to assess because of the large variation in cost per ultrasound examination. On the one hand, missed hip dysplasia may lead to surgery, early osteoarthritis, and the risk of leukemia from high doses of radiation exposure during treatment. On the other hand, the high positive rate generated by ultrasound screening programs can lead to a high treatment rate that is considered to be beyond normal, a condition that seems to occur more often in infants with mild acetabular dysplasia or immaturity.
Although overtreatment does not imply an increase in surgical rates, invasive treatment with Pavlik braces or splints may lead to ischemic necrosis of the femoral head. This lesion originates from a decrease in the blood supply to the internal rotor femoral artery (which is the main source of blood supply to the developing femoral head) and is caused by compression of blood vessels by the iliopsoas tendon and other para-articular structures during the implementation of excessive abduction fixation. Energy Doppler imaging and spectral Doppler analysis can show arterial blood flow signals within the femoral cartilage. With a progressive increase in abduction angle, a loss of femoral blood flow signal can be observed.
In the context of DDH treatment, energy Doppler imaging can help reduce the risk of ischemic necrosis of the femoral head by making predictions in children with undetectable blood flow at 60° of hip abduction and by showing that blood flow signals remain after orthopedic device use. Recently, a study found that alpha angle and blood flow spectral resistance index are directly proportional. During follow-up, ultrasound can help monitor the position of the femoral head relative to the acetabulum: this will be used to determine the length of treatment and to evaluate joint stability in order to adjust the brace accordingly.