Congenital Dislocation of the Hip

This article reviews changes that have occurred in the understanding and treatment of congenital dislocation of the hip (CDH). It is now recognized that CDH is a spectrum of diseases with differing etiologies, pathologies, and natural histories that, as a result, require different treatment approaches. The etiology of CDH involves genetic, hormonal, mechanical, and environmental influences. This recognition of mechanical factors that predispose to CDH has allowed the identification of high-risk patients. Screening for CDH in the newborn has resulted in many more patients being diagnosed earlier. Ultrasonography has added a new dimension to CDH diagnosis, since it has demonstrated abnormal hips in otherwise normal children. In the newborn, the use of ultrasonography may be the single most important recent contribution to the study of this disease. The natural history of CDH is well-known: The disease often can lead to early coxarthrosis. However, avascular necrosis is a complication of treatment and results in a hip that may be worse than one not treated at all. Treatment of CDH includes gentle closed reduction in the majority of patients. In patients under six months of age, this can usually be achieved with the Pavlik harness. Up to the age of three years, gentle manipulative closed reduction after traction is successful in the majority of pa-

tients. The trend toward the use of home traction and away from inpatient skeletal traction has been documented by many centers to be successful. Open reduction is becoming infrequently required and has specific indications. Still, this reduction must be gentle and may follow traction in the younger patient or femoral shortening in the older patient if it is not possible or desirable to lower the femoral head to the level of the acetabulum. Secondary procedures, including the Salter or Pemberton osteotomies, are indicated in the older patient for persistent dysplasia or subluxation when the acetabulum fails to respond to the concentric reduction of the femoral head into the acetabulum. The Chiari or shelf salvage procedures are indicated when a stable concentric reduction cannot be achieved.
In the study of disease, it is important to accurately define the disease studied. Probably one of the most significant advances in the study of congenital dislocation of the hip (CDH) has been the recognition that it is a spectrum of diseases.' Everyone would agree that, for the diagnosis of a true dislocation, Ortolani's sign of reduction should be present initially, after a period of traction, or at the time of open reduction. However, as GrafI6 has confirmed by ultrasound technique, not all abnormal newborn hips demonstrate complete dislocation. Furthermore, there may be a subset of this condition that appears as acetabular dysplasia and subluxation in the older patient.' Therefore, CDH should be recognized not as a single entity but as a spectrum of diseases with different etiologies, varying pathologies, and differing Clinical Orthopaedics and Related Research natural histories that require different treatment modalities. ETIOLOGY Genetic, hormonal, mechanical, and environmental influences predispose the hip joint to dislocation. The incidence of CDH in the United States is approximately ten cases per 1000 live births. The dislocation rate among second children born to parents of a child with CDH approximates 22 to 50 cases per 1000 live births.' Mechanical factors present in utero that predispose to CDH include breech presentation (even with cesarian section), congenital recurvation or dislocation of the knee, muscular torticollis, and metatarsus adductus. Physiologic factors to be considered include the role of maternal estrogens and the known higher incidence of CDH in females. The ratio of the occurrence of CDH in girls to boys is 6: 1.

PATHOLOGY
In all newborns there is normally laxity in all joints, including the hip joint capsule. In contrast, in the older child with a persistent dislocation, both soft tissue and bony parts are distorted. Ultrasonography has demonstrated soft-tissue structural abnormalities present in the newborn hip." The acetabular labrum may be folded into the posterior superior lip of the acetabulum as a minor invagination, or it may form an almost complete dia~hragm.~' Fat and fibrous tissue in the depth of the acetabulum may be enlarged (pulvinate), obstructing reentry of the femoral head. The capsule may be constricted between the acetabulum and the femoral head, producing an hourglass deformity. The iliopsoas tendon may indent the inferior aspect of the capsule and, together with the intact transverse acetabular ligament, act as a bamer to the full reduction of the femoral head. Very early, the femoral head may lose its spherical shape and become flattened. In the persistently dislocated or subluxated hip, femoral anteversion fails to decline.28 In the older child, the delay in the ossification of the roof of the acetabulum is recognized roentgenographically as acetabular dysplasia with an increase in the acetabular index.43 Acetabular dysplasia makes evaluation of the center-edge angle difficult, but it is absolutely necessary to distinguish between dysplasia and subluxation. Dynamic arthrography will help make that distinction.

NATURAL HISTORY
The minimal disability of many untreated complete dislocations of the hip, particularly bilateral dislocations, is well-known. However, the early disability is minor relative to hips with residual subluxation or complicated by avascular necrosis, both of which result in an early c o x a r t h r o~i s .~~~~l The precise mechanism of avascular necrosis is not known, but it is only seen with treatment. Therefore, treatment in which avascular necrosis is a significant complication does not necessarily alter the natural history more favorably than no treatment at all. Unfortunately, studies with initial excellent roentgenographic results have demonstrated early symptomatic degenerative changes on longer follow-up ob~ervation.~' In a study of untreated patients, Wedge and Wasylenko4' reported significant disability in untreated adults with CDH. Pain appears to be a minor early problem; but severe fatigue, unsightly unstable gait, postural scoliosis, valgus deformity of the ipsilateral knee, and low-back pain are reported.
It is estimated that congenital subluxation is the cause of 20%-50% of coxarthrosi~.~~ Disabling coxarthrosis may, in fact, be inevitable in subluxation. This is true in the dysplastic hip but to a lesser extent. The difficulty with dysplasia is that its recognition is usually fortuitous, since it is usually asymptomatic and is often found contralateral to a dislocated or subluxated hip.6 It is therefore important to observe the apparently normal side. Dysplasia resolves spontaneously in one-half of hips,35 but which hips will develop normally is difficult to predict.

TREATMENT
The key to the treatment of congenital hip disease is early diagnosis. Based on this premise, screening programs have been developed. However, these programs in many areas of the world have failed to reduce the incidence of late The diagnosis of CDH in the neonate is based or suspected on the clinical examination. The classic jerk sign of reduction is unusual in the newborn, though it may be present. The provocative maneuver of Barlow2 is subluxation or dislocation of the hip when the extremity is adducted and downward pressure is exerted. The leg is then abducted in the manner of Ortolani, and the hip reduces.
Roentgenograms play a minor role in the diagnosis of CDH in the newborn. In contrast, ultrasound is an increasingly reliable and accurate method of diagnosis of suspected hip subluxations and dislocation^.^^" The cartilaginous femoral head in its relative position to the triradiate cartilage is well-visualized, as are deformation of the capsular structures and the cartilageous acetabulum. Sensitivity has been reported to be approximately 90%. Ultrasonography promises to be the most helpful adjunct in the diagnosis of congenital hip disease in the infant.
Under the age of one year, most patients can be managed by nonoperative means. There is evidence to support attempted closed treatment for children under age three years. 48 The Pavlik harness is the mainstay in treatment of congenital dislocation and subluxation up to six months of age. Correctly applied, the incidence of avascular necrosis has been reported to be as low as zero, al-though some series have reported higher rates with forced abduction.2' The principles of treatment include obtaining and documenting a concentric reduction through the use of flexion. Failure rates of 17% have been reported and are often associated with bilateral CDH, late dislocation, and improper use of the Pavlik harness. Although the harness may be properly applied, reduction may be blocked by obstructing r n a t e~i a l .~~,~~.~'

CLOSED REDUCTION
Probably, the single most important principle in the treatment of the child who fails the Pavlik or is too old for its use is gentle reduction after the use of traction. In a review of 20 years' experience, Gage and Winter,14 through the use of preoperative traction, achieved normal hips in 58.6% of cases when the femoral metaphysis was brought to the level of the triradiate cartilage (traction station zero). When Shenton's line was restored (traction station two), the result was a normal hip 95.5% of the time.14 There is a strong trend away from inpatient skeletal traction based on reports of the successful use of skin traction reported that can often be used as a home program.20 Following traction, which may be done in the hospital but can usually be carried out at home, gentle closed reduction under general anesthesia and the application of a hip spica cast in the human position is the treatment protocol for children six months of age and for many patients up to the age of three years. Closed reduction was successfully obtained and maintained in 38 of 51 hips in children between the ages of one and three years. In the absence of subluxation, acetabular development rapidly occurred and continued for as long as six years postreduct i~n .~~ The remarkable potential for acetabular recovery has been documented after closed reduction by Harris et a1.,I9 Lindstrom el and many others.

Clinical Orthopaedice and Related Research
All maneuvers must be done as gently as possible and are better thought of as positioning rather than forced reduction. Tightness in the adductors is an indication for a subcutaneous adductor tenotomy. Since the patient is anesthetized, adductor tightness may not be appreciated until after the patient awakens. Therefore, adductor tenotomy is recommended in almost all children. It may improve the safe zone, i.e., that zone after a reduction between the extremes of forced abduction and redislocation. The importance of flexion in order to achieve a stable reduction is increasingly recognized. Confirmation of the reduction by arthrography is paramount. While the femoral head may be directed to the triradiate cartilage, it may require some time to settle in; a repeat arthrogram within weeks must demonstrate the reduction is complete. Pseudoreduction may be clinically indistinguishable from a true reduction without arthrography. The labrum may be inverted and lead to late dysplasia. The proper, but expeditious, application of the hip spica cast while the dye is still present may document that the reduction has not been lost during the application of the cast. A computed tomographic scan should be obtained if there is any doubt as to the adequacy of the reduction. There may be a role for magnetic resonance imaging of the hip, since that modality may give not only structural information as to the adequacy of the reduction but also information as to the biologic status of the hip postreduction.

OPEN REDUCTION
Open reduction is indicated when (1) the femoral head persistently lies above the triradiate cartilage, (2) the arc of reduction redislocation is less than 25" after adductor tenotomy, (3) after adductor tenotomy, the femoral head will not enter the acetabulum, (4) the femoral head is still laterally placed in the acetabulum after four weeks of partial reduction, and/or ( 5 ) a previous reduction has failed. For most patients, an open reduction is the primary procedure; femoral or pelvic osteotomy is reserved as a secondary measure. While combined procedures may be indicated for the more complex hips, there is a higher redislocation rate following combined procedures. The medial adductor approach (Ludloff approach) is controversial. Ferguson13 found no avascular necrosis in 125 patients treated by a posteromedial approach to the adductor longus tendon. W e i n~t e i n~~ proceeds anterior to the pectineus, and Staheli39 prefers the medial approach in arthrogryposis. Because of significant avascular necrosis, in excess of 50% in some series, the medial approach remains controversial.22 The advantage of the medial approach is release of the often obstructing transverse acetabular ligament. However, eversion of the labrum may be difficult, and it is generally recognized that excision of the labrum should be avoided.
The Smith-Petersen approach is the standard technique to the open reduction of the congenitally dislocated hip. The skin incision should be placed below the iliac crest and can be continued in a line parallel to the inguinal ligament. Psoas tendon tenotomy is in keeping with the principle of a gentle reduction. Fibrofatty tissue from the acetabulum should be removed; the transverse acetabular ligament should be palpated, outlined, transected, and spread to allow a pocket to be developed inferiorly.
The infolding of the acetabular labrum (limbus) may be a critical obstruction to concentric red~ction.'~ If possible, it should be everted or flipped over the femoral head. Occasionally, this may require one or two radial cuts to allow passage of the femoral head under the labrum. A T-shaped incision in the capsule of the hip joint allows removal of any impediment to reduction.37 After reduction, the capsule is closed by overlapping or excision of redundant portions. A compli-cation of open reduction, a stiff hip, is best avoided if cast immobilization is not continued past six to eight weeks.

FEMORAL SHORTENING
Femoral shortening combined with open reduction markedly reduces the incidence of avascular necrosis in patients older than three years. Currently, the trend is to avoid the use of preoperative skeletal traction. Though advocated by Crego," skeletal traction in the older patient is associated with avascular necrosis. Schoenecker and S~a k e r~~ reported a 54% incidence of asceptic necrosis and a 3 I % redislocation rate with the use of skeletal traction in patients older than three years of age. Femoral shortening is reported by several investigators to be an aid in facilitating reduction and decreasing the potential of complications. 1s*24.253

SECONDARY PROCEDURES
In general, it is wise to achieve a stable open reduction before performing secondary procedures, and it is usually not necessary to perform either an acetabular or femoral procedure in order to achieve a stable reduction. Biologic remodeling can then be observed, and the remarkable ability of the acetabulum to develop can be observed. Subluxation should not be tolerated. Subluxation at times may not be suspected unless the roentgenograms are obtained with the patient standing.
The center-edge angle should be at least 20" and preferably 25" to ensure normal seating of the femoral head. Acetabular dysplasia produces no problems and can be observed for several years. When there is persistent dysplasia or subluxation, secondary procedures are absolutely indicated. In general, one aspect of the joint is usually more deformed than the other; it is best to treat the more serious deformity. As a rule, both the pelvis and the proximal femur lose their abil-ity to remodel after seven to eight years of age. After this age, procedures to improve both the femoral and acetabular components of the deformity must be considered. Before the age of four years, changing the configuration of the proximal femur to redirect the force on the acetabulum facilitates acetabular improvement as long as there is not a significant instability of the hip On the acetabular side, both the Salter and Pemberton osteotomies emphasize improvement of anterolateral acetabular deficiency. In older patients with limited mobility of the symphysis pubis, the Steele triple innominate osteotomy is more effective. The Sutherland osteotomy rarely offers a real advantage over the Salter or Steele osteotomies. The dial (Eppright) or even more difficult Wagner redirectional osteotomies are noted for completeness but are technically demanding and have limited, albeit specific, indications. Salvage procedures for the adolescent and young adult are limited. However, since total hip arthroplasty has not yielded good long-term results in young patients, the only alternatives are salvage procedures. The shelf operation may increase the volume of the acetabulum and provide a greater surface area through which to distribute forces. In many patients, coxa magna as well as subluxation of the hip result in lateralization of the moment arm for the hip and decrease the efficiency of the hip biomechanics. Although theoretically the Chiari osteotomy moves the center of hip rotation medially (medialization), this has not been found to any significant degree.4 This osteotomy does increase the surface area for the contact of the femoral head against the side of the pelvis, and it may decrease the eccentricity of the hip. It should not be used in young patients in whom concentric reduction is possible.