Proper growth and development of the acetabulum is due to the presence of the femoral head in adequate contact and without stress within it. In other words, the acetabulum needs the femoral head for its development. DDH is the result of the disruption of this relationship. It is estimated that in 98% of the cases it results from a late intrauterine event of a previously well formed hip due to persistent forces. In 2% of cases it results from an early intrauterine event secondary to a congenital neuromuscular disorder. The term used for the latter type of hip is "teratologic". In a small number of cases DDH occurs later, beyond the neonatal period, despite initial normal clinical, and sometimes normal ultrasound, examinations. ¹

The hip develops from a single block of mesoderm, and at the beginning of the fetal period (8 weeks since conception) all the structures of the joint are already in place (figure 1). ² At birth, the acetabulum has a smaller bony component and a larger cartilaginous one, ³ and the femoral head coverage by the acetabulum is less than at any other time during the fetal or postnatal periods. ⁴ This situation results in a 6-week postnatal period in which the acetabulum is highly susceptible to modeling; it is slightly less susceptible from 6 to 12 weeks and very slightly susceptible after 16 weeks. If the femoral head is in a normal position within the acetabulum the end result will be a normal hip. This is the key period for treatment of DDH. If the femoral head is in an abnormal relationship and this is not corrected, the end result will be a permanently dysplastic hip. Early diagnosis is, therefore, pivotal. However, it is important to know that with minor ultrasound findings, 78% of the hips will spontaneously become normal by the fourth week and 90% by the ninth week. ⁵ Therefore, a conservative attitude in the interpretation of the ultrasound findings is warranted.

The consideration of some facts should help us understand, in a particular case, the possibilities of a successful ultrasound examination and a positive diagnosis.

The first problem encountered in examination of the infant hip is technical and related to visualization of the structures. The femoral head begins ossifying between the second and eighth month of life in males, and earlier in females (figure 2). On ultrasound, visualization of the neovascularity of the future ossification center begins weeks before the radiological findings. As the size of the ossification center progressively enlarges it will, at some point, obscure the deeper acetabulum, making ultrasound examination impossible. The age of occurrence of this event is variable, mostly occurring in the second semester of life. If the patient is in that age range we might attempt the ultrasound examination first and move on to a radiograph in case of failure. Some authors are more dogmatic and recommend ultrasound only up to 4 months of age, with radiography thereafter. ⁶

The second problem is deciding the timing of the ultrasound examination on the basis of the clinical history and results of physical examination. A newborn with an abnormal physical examination will benefit from the ultrasound examination if performed promptly within the first 2 weeks of life. Newborns with a risk factor by history but with a normal physical examination can wait for 4 to 6 weeks. In this way, the normal instability of the hip of the first weeks of life will be avoided. Similarly, many of the morphologically immature hips will progress towards normality, given time. ⁷

Articles from different countries report a different incidence of DDH in all of its forms. For example, Australia reports a 1% incidence, Netherlands 3.7%, Poland 3.9%, Israel 5.9%, Austria 6.57%, and Norway 16.9%. The reported average of hip dislocation is 0.6 per thousand newborns, whereas for indigenous North Americans and Lapps it can be as high as 25 to 50 per thousand. On the other hand, Chinese and Black Africans have a negligible incidence, but American Blacks are affected. The difference is most likely genetic, and Blacks have been found to have deeper acetabula at birth. Additionally, environmental factors cannot be ignored. It is interesting to note
that in the group with a high incidence, mothers traditionally swaddle their infants with the hips in extension, or strap them to a cradle board; both result in unphysiological hyperextension of the hips. On the contrary, in the groups with a low incidence of the problem, mothers carry their infants against their waist with the children's legs in flexion and abduction, which is more natural. ⁸ On the same venue, in Japan after the institution of a national program to discourage the swaddling of infants, the incidence of DDH dropped from 3.5 to 0.2%. ⁹ There is a higher incidence of DDH in Central Europe and in the South American Andean countries.

A higher winter incidence of DDH, in the range of 1.5% as compared to that of summer (1%), has been reported. ¹⁰

Cases of families with several members involved by DDH are well known. There is a reported familial incidence in 20% of patients. ¹¹

The risk of DDH for a female fetus in breech presentation with a maternal history of the disease is 1 in 15; the risk for those with a positive maternal history but a presentation other than breech is 1 in 25. ¹² A newborn with a sibling affected by the dysplasia carries a risk of 6%, 12% if one parent is affected, and 36% if both a sibling and a parent are. ¹³

DDH is from 4 to 8 times more frequent in females than in males. ¹⁴ This is thought to be possibly related to an increased level of circulating estrogen observed in affected newborns, to which females are more sensitive. Estrogenic action results in the blockage of maturation of collagen, thus affecting the development of the acetabulum. ¹⁵ However, elevated levels of estrogens were not confirmed by other researchers. Similarly, the pregnancy hormone relaxin has been found in variable levels in newborns with DDH, representing another possible causative factor. ¹⁶

The development of a fetus requires enough room to move freely without inordinate tension applied. This is as valid for the proper development of the extremities as for the lungs. When ample space is not available, abnormalities such as DDH can develop. For example, oligohydramnios severely restricts motion and places stress on the extremities, which are placed in hyperflexion and/or hyperextension for prolonged periods of time (figure 3).

Additionally, first pregnancies find unstretched maternal abdominal and uterine walls which also may limit free movement. Sixty percent of patients with DDH occur in primiparous mothers. Breech presentation results in increased tension and hyperflexion of the hips due to their location within the inelastic maternal pelvis and the impossibility of free active motion. This is observed in 30 to 50% of patients with DDH, whereas only 2 to 4% of deliveries are in the breech presentation. Most fetuses in breech presentation lie with the left lower extremity on the maternal spine which forces its adduction and limits its motion. This is thought to be responsible for the increased incidence of left hip involvement that is seen in 80% of these patients.

Bilateral involvement is described in 25%. ^17,18 High birth weight also is associated with an increased incidence of DDH. ^l9 Twin fetuses have a similar incidence of DDH to that of singletons. ²⁰ The mode of delivery, vaginal or by Cesarean section, does not seem to affect the likelihood of DDH. ²¹

DDH has an association with other postural and non-postural musculoskeletal abnormalities. It is found in 2% of patients with club foot and metatarsus varus and in 20% of those with torticollis. ¹⁴ Also, it has been seen with scoliosis, head and facial deformities, generalized joint laxity, spina bifida, sacral agenesis, myelodysplasia, arthrogryposis multiplex, and cardiac and renal abnormalities (due to secondary oligohydramnios), as well as other syndromes.

We might receive a referral for ultrasound on the basis of abnormal physical findings. On observation, an infant with a dislocated hip might show a shortened lower limb and redundant skin folds in the thigh due to an apparent excess of skin. Both findings are secondary to an upward displacement of the dislocated femoral head. These findings are more likely to be seen in older infants, not in the newborn. ²²

By performing different maneuvers we may find a limitation of abduction on the affected side as compared to the opposite side. The Barlow's maneuver, performed with hip and knee flexion, attempts to dislocate a well placed femoral head (figure 4). With the limb in adduction, a gentle posterior push, like a piston, is made. When positive (subluxable or dislocatable hips), we refer to the hip as "unstable". It should be noted that a mild displacement of a few millimeters is normal in the first 2 weeks of life due to circulating humoral factors. Clinically, instability is found in 1 to 3% of all newborns.

The Ortolani's maneuver attempts
to reduce a dislocated hip (figure 5).
It is performed with flexion of the hip and knee. While the extremity is being abducted, an anterior and sustained push to the thigh is applied from the posterior aspect, attempting to relocate the posteriorly dislocated femoral head. On the basis of the results,
we refer to the dislocated hip as "reducible" or "non-reducible". It should be noted that the sensitivity of the clinical Ortolani's and Barlow's maneuvers has been reported by several authors, ranging from 10 to 34.4%, and their specificity has been found to be from 84.3 to 99%. ²³ False negative results of the Ortolani's maneuver can occur in newborns with extreme capsular laxity in which the dislocated femoral head can be fully abducted without actually reducing it. It also can occur in older infants with a longstanding dislocation in which the hip muscles become contracted and shortened, trapping the femoral head outside the acetabulum. ¹⁴ The described maneuvers have been incorporated into the ultrasound examination, improving their sensitivity and specificity.

At our institution we sometimes get referrals because of audible high pitch "clicks" observed during the performance of the described clinical maneuvers. In general, this finding results from joint capsule and tendon stretching and snapping, and has no pathological significance. Occasionally, though, we might find a positive finding on ultrasound. ²⁴

The anatomy of the infant hip is illustrated in figures 6 and 7. Two techniques for the ultrasound examination of this area are widely in use. Both use a lateral approach with the infant supine or in the lateral decubitus position. The first reported technique is based on the morphology of the acetabulum. It was described by Graf in 1980 ²⁵ and consists of a single coronal image through the deepest portion of the acetabulum (figure 8). Its key is to correctly identify this "standard plane". For that, we have to place the iliac bone parallel to the surface of the transducer. This is achieved by gently sliding the transducer in the anteroposterior direction. The final step is to obtain a sharp definition of the lower end of the bony acetabulum. This is achieved by slightly rotating the transducer. If these parameters are obtained correctly, we should be in view of the entire acetabulum, including its cartilaginous portion and labrum.

On the described image, the slope of the acetabulum (alpha angle) is measured with respect to the iliac line. An angle of 60 degrees or more is normal; from 50 to 60 degrees is considered physiologic before 3 months of age, but needs to be followed for observation. Values under 50 degrees are abnormal at any age. A second angle (beta) is drawn between the iliac line and a line drawn from the labrum to the transition point between the iliac bone and the bony acetabulum. This measurement is indicative of the acetabular cartilaginous roof coverage and is secondary in significance to the alpha angle. The smaller the angle the less the cartilaginous coverage due to a better acetabular bony containment of the femoral head.

Graf's classification of DDH is based on these 2 angles. In this technique the position of the femoral head is not considered, based on the premise that its position (normal, subluxed, or dislocated) will be reflected by the morphology of the evaluated acetabulum. Figures 9 and 10 are examples of a subluxed and a dislocated hip, respectively.

The second ultrasound technique used in the diagnosis of DDH is dynamic and was described by Harcke et al in 1984. ²⁶ This technique incorporates the use of real time. The examination is performed in the coronal, similar to Graf's technique, and axial planes. However, two key points differentiate it from Graf's. First, it takes into consideration the position of the femoral head. Second, it incorporates the Barlow's maneuver in both imaging planes in trying to demonstrate instability (subluxable, dislocatable hips) (figure 11). Sonographically, mild instability is found in all newborns for the first few days, with spontaneous resolution in normal cases. When finding a dislocated hip, the Ortolani's maneuver should be performed to check for reducibility.

Different institutions use one method or the other, or a combination of the two, as recommended at a symposium on the subject held in Maryland in 1993. The recommended protocol of a "dynamic standard minimum examination" asks for the patient to be placed in a supine or lateral decubitus position. Scanning is performed in the coronal plane with the hips extended or flexed. Stress views and angle measurements are optional. In the axial plane the thighs are in 90 degrees flexion, and images are obtained with and without stress. ²⁷

A third method, which is complementary to the other two described above, is based on the measurement of the acetabular coverage of the femoral head (figure 12). Described by Morin et al in 1985, ²⁸ this maneuver is based on the radiographic migration percentage (MP). It is calculated by the equation ( d /D) * 100. On the above described "standard plane" of the coronal projection, d is the distance from the iliac line to the medial aspect of the femoral head; D is the maximum diameter of the femoral head. An acetabular coverage of the femoral head of 58% or more is normal, and below 33% is abnormal (subluxation). Intermediate values are indeterminate. This measurement is of no use in dislocation because the acetabular/femoral head relationship is lost. Values are unrelated to patient's age. It is an effective method because it is intuitive. At a glance, we can see if more than half of the femoral head is covered by the acetabulum, which tells us a normal
hip is present. The less the coverage the more immature the acetabulum.

Finally, it is important to understand the relationship between hip morphology and instability in newborns. In one study, out of 80 morphologically dysplastic hips, 91% were sonographically unstable or dislocated, and out of 142 sonographically unstable hips, 49% were morphologically normal or physiologically immature. ²⁹ Therefore, treatment should be indicated on the basis of morphology and/or instability. AR