I. Developmental Dysplasia of the Hip
a. Developmental dysplasia of the hip (DDH) refers to the complete spectrum of pathologic conditions involving the developing hip, ranging from acetabular dysplasia to hip subluxation to irreducible hip dislocation.
b. Teratologic dislocation of the hip occurs in utero and is irreducible on neonatal examination.
i. Pseudoacetabulum usually is present.
ii. This condition always accompanies other congenital anomalies or neuromuscular conditions, most frequently arthrogryposis and myelomeningocele.
a. DDH is the most common disorder of the hip in children. One in 1,000 children (0.1%) is born with a dislocated hip; 10 in 1,000 children (1%) are born with hip subluxation or dysplasia.
b. 80% of affected children are female.
c. The left hip is more commonly involved (60%).
d. DDH occurs more commonly in Native Americans and Laplanders; DDH is rarely seen in African-Americans.
1. Etiology—The exact cause is largely unknown but is thought to be multifactorial (genetic, hormonal, and mechanical).
2. Risk factors
a. DDH occurs more commonly in females and firstborns, and with breech presentation (30% to 50%).
b. DDH is commonly associated with intrauterine "packaging" problems, such as prematurity, oligohydramnios, congenital dislocation of the knee, congenital muscular torticollis (8% to 17% coexistence), and metatarsus adductus (0% to 10% coexistence).
c. Family history is a strong risk factor.
i. No parent involvement: 6% risk for DDH
ii. At least one parent involved: 12% risk
iii. Parent and sibling involved: 36% risk
1. Clinical presentation—The clinical presentation varies with age.
a. In the neonatal period, instability of the hip is the key clinical finding.
b. Hip clicks are nonspecific physical findings.
c. Because asymmetric skin folds are common in children with normal hips, children with such asymmetry have a high rate of false positives.
d. In infants older than 6 months, limitation of motion and apparent limb shortening are common findings.
e. In toddlers, in addition to restricted motion and limb-length inequalities, a limp or waddling gait may be appreciated.
f. In adolescents, all the above findings may be present in addition to fatigue and pain in the hip, thigh, or knee.
2. Physical examination—Accuracy of the physical examination requires that the child be relaxed.
a. The Galeazzi (or Allis) test is positive only in unilateral severe subluxations or dislocations. In the Galeazzi test, the hips are flexed to 90°; the test is positive if one knee (the involved side) is lower than the other.
b. The Barlow test is performed by applying a posterolateral force to the extremity with the hip in a flexed and adducted position (
Figure 1). The test is positive if the hip subluxates or even dislocates.
c. The Ortolani test is performed by abducting and lifting the proximal femur anteriorly (
[Figure 1. The Barlow test is positive if the hip subluxates or dislocates.]
The test is positive if the dislocated hip is reducible.
d. Range of motion (ROM) testing of the hip is important, with decrease in abduction as the most sensitive test for DDH. Remember, however, that ROM will be normal in children younger than 6 months because contractures will not yet have developed.
3. Diagnostic tests
i. In the first 4 months of life, plain radiographic evaluation is unreliable because the femoral epiphysis has not yet ossified.
ii. Ultrasonography of the infant hip (before the appearance of the proximal femoral ossific center) is useful in confirming a diagnosis of DDH and also in documenting reducibility and stability of the hip in an infant undergoing treatment with a Pavlik harness or brace.
iii. Reference parameters (
(a) At age 4 to 6 weeks, a normal α angle is >60°; a normal β angle is <55°. (The α angle is the angle formed by a vertical reference line through the iliac bone and tangential to the osseous roof of the acetabulum; the β angle is the angle formed by a line drawn through the labrum intersecting the iliac reference line—ie, it represents the cartilaginous roof of the acetabulum.)
[Figure 2. The Ortolani test is positive if the dislocated hip is reducible.]
(b) The amount of femoral head coverage should be >50%.
b. Plain radiographs—
Figure 4 shows reference lines and angles for the AP view of the pelvis (
i. The Hilgenreiner line is a line drawn horizontally through each triradiate cartilage of the pelvis.
ii. The Perkin line is drawn perpendicular to the Hilgenreiner line at the lateral edge of the acetabulum.
iii. The Shenton line is a continuous arch drawn along the medial border of the femoral neck and superior border of the obturator foramen.
iv. The acetabular index is the angle formed by an oblique line (through the outer edge of the acetabulum and triradiate cartilage) and the Hilgenreiner line.
(a) In the newborn, a normal value averages 27.5°.
[Figure 3. Ultrasound images of a normal hip and a hip with DDH. A, Ultrasound image of a normal hip. B, The same ultrasound image with the α and β angles drawn. In a normal hip, femoral head coverage should be greater than 50%. The α angle should be greater than 60°. C, Ultrasound image of a dysplastic hip reveals approximately 30% femoral head coverage, an α angle of 50°, a β angle of 90°, and an echogenic labrum.]
[Figure 4. Reference lines and angles used in the evaluation of DDH.]
(b) By 24 months of age, the acetabular index decreases to 21°.
v. The center-edge angle of Wiberg is the angle formed by a vertical line through the center of the femoral head and perpendicular to the Hilgenreiner line and an oblique line through the outer edge of the acetabulum and center of the femoral head (
(a) The center-edge angle is reliable only in patients older than 5 years.
(b) A center-edge angle <20° is considered abnormal.
c. Arthrography of the hip—Useful in confirming the acceptability of a closed reduction and in diagnosing the blocks to reduction, such as capsular narrowing and labral hypertrophy.
[Figure 5. AP plain radiograph of the pelvis demonstrates a dislocated right hip. The proximal femoral ossific nucleus is not yet present. Also note that the Shenton line is disrupted.]
d. CT—The standard for confirming acceptable reduction for a patient in a spica cast following closed or open procedures.
e. Magnetic resonance imaging
i. MRI can also be used to confirm acceptable reduction of the hip following closed or open procedures.
ii. MRI can also be useful in an older patient with suspected labral pathology.
4. Neonatal screening
a. Clinical screening (thorough history taking and physical examination) of all newborn infants is necessary.
[Figure 6. The center-edge angle of Wiberg is formed between two lines passing through the center of the femoral head; one extends to the lateral edge of the sourcil (A), and the other is a perpendicular to the interteardrop line (B).]
b. Ultrasound screening of all newborn hips appears unnecessary.
i. Routine ultrasound screening should be performed for infants with risk factors for the condition.
ii. Screening by ultrasound should be delayed until age 4 to 6 weeks (or corrected age for premature infants) because ultrasononography is associated with poor specificity in the initial newborn period (4 to 6 weeks).
D. Treatment—Based on the age of the child, stability of the hip (unstable versus dislocated hip), and severity of acetabular dysplasia (
1. Dysplastic hip in neonate through 6 months of age
a. In a child with an abnormal abnormal α angle on ultrasound or with an unstable hip (subluxatable hip on examination), initial treatment usually includes a Pavlik harness.
i. Proper positioning of the Pavlik harness is critical.
(a) The hips should be flexed to 100° with mild abduction (two to three finger breadths between knees when knees are flexed and adducted).
(b) Excessive flexion should be avoided to lower the risk of femoral nerve palsy.
(c) Excessive abduction should be avoided to lower the risk of osteonecrosis; osteonecrosis can occur in both the normal and dysplastic hip.
ii. The child can be weaned from the Pavlik harness over a 3- to 4-week period when ultrasound parameters become normal.
iii. Success rates for Pavlik harness treatment in this setting have been reported at >90%.
iv. The recurrence rate is 10%; therefore, follow-up evaluation until maturity is necessary.
b. In a relatively large child or in a child older than age 6 months with a dysplastic hip or with hip subluxation, a fixed abduction orthosis or spica casting is an option. Pavlik harness treatment is ineffective in this setting because the child generally "overpowers" the brace.
[Table 1. Management of Developmental Dysplasia of the Hip]
2. Dislocated hip in neonate through 6 months of age
a. If the hip is Ortolani-positive, Pavlik harness treatment is initiated.
i. Frequent (every 1 to 2 weeks) reexamination (clinical plus ultrasound) is necessary to ensure that the hip is reduced.
ii. Once the hip becomes stable, treatment is the same as the protocol described above for treatment of dysplastic hip.
iii. Success rate (ie, hip becomes reduced) is reported to be 85%.
iv. The risk of osteonecrosis is low (<5%), especially if treatment is initiated early (ie, in the first 3 months of life).
b. If the hip is Ortolani-negative, initial treatment still includes a Pavlik harness. If reduction is not achieved by 3 to 4 weeks (confirmed by ultrasound), however, Pavlik harness treatment is abandoned, and closed reduction is necessary.
i. Pavlik harness treatment should be discontinued if the dislocated hip does not relocate within 3 to 4 weeks, to avoid Pavlik harness disease.
ii. In Pavlik harness disease, the femoral head sits up against the edge of the acetabulum and worsens the acetabular dysplasia, particularly the posterolateral rim.
3. Dislocated hip in children 6 to 18 months of age
a. Closed reduction is the preferred method of treatment in children age 18 months or younger.
b. Secondary femoral or acetabular procedures are rarely necessary in this age group.
c. The evidence for preliminary traction is equivocal. Given this and the possible complications of skin slough and leg ischemia, most centers have abandoned preliminary traction.
d. Closed reduction is performed under general anesthesia.
i. Adductor tenotomy frequently is necessary.
ii. Hip arthrography is used intraoperatively to confirm adequacy of reduction.
iii. There should be <5 mm of medial contrast pooling between the femoral head and acetabulum.
iv. The safe and stable zones for abduction/adduction, flexion/extension, and internal/external rotation should be established.
v. A spica cast is applied with the hip maintained in the "human position" (hip flexion of 90° to 100° and abduction). Hip abduction should be <60° to minimize the risk of osteonecrosis.
vi. The reduction of the hip in the cast must be confirmed.
(a) Presently, CT is the standard.
(b) MRI is used in some centers to confirm adequacy of reduction of the hip.
vii. Cast immobilization is continued for 3 to 4 months.
(a) Cast change is sometimes necessary.
(b) Removable abduction brace is used afterwards until the acetabulum normalizes.
4. Dislocated hip in children >18 months of age
a. Open reduction is the preferred treatment.
i. Unilateral dislocation: Surgical treatment is generally indicated in children up to 8 years of age with a unilateral dislocation. After 8 years of age, the risks of surgery are felt to outweigh the benefits.
ii. Bilateral dislocation: The upper age limit for surgical treatment in these children is typically 5 to 6 years.
b. Femoral shortening is indicated in children with significantly high-riding dislocations.
i. This is necessary in most, but not all, children ≥ 2 years of age.
ii. Pelvic osteotomy is needed for significant acetabular dysplasia (typical in children > 18 to 24 months old). If a pelvic osteotomy is performed in this age group at the time of initial surgery, the rate of reoperation is reduced significantly.
5. Open reduction
a. Open reduction is indicated if concentric closed reduction cannot be achieved or when excessive abduction (>60°) is required to maintain reduction.
b. The goal of open reduction is to remove the obstacles to reduction and/or safely increase its stability (
i. Impediments to congruent reduction are the iliopsoas, hip adductors, capsule, ligamentum teres, pulvinar, and transverse acetabular ligament. An infolded labrum may be an impediment in some cases.
ii. The most commonly used approaches are anterior, anteromedial, and medial (
6. Secondary procedures
a. Secondary femoral and/or pelvic procedures are frequently necessary in children older than 2 years of age to achieve and maintain concentric reduction and minimize the risk of osteonecrosis.
b. Femoral osteotomy
i. Femoral osteotomy provides shortening (to decrease pressure on the femoral head, thereby minimizing the risk of osteonecrosis), derotation (external rotation to address the abnormally high femoral anteversion in DDH), and varus.
ii. Avoid excessive varus, because the greater trochanter can impinge against the acetabulum and prevent concentric reduction.
c. Pelvic osteotomy
i. Indications for pelvic osteotomy include persistent acetabular dysplasia and hip instability.
ii. There is considerable variability in clinical practice with regard to pelvic osteotomy in children > 2 years of age.
iii. The two general types of pelvic osteotomy are reconstructive and salvage (
(a) Reconstructive osteotomies redirect or reshape the roof of the acetabulum with its normal hyaline cartilage into a more appropriate weight-bearing position. A prerequisite to a reconstructive pelvic osteotomy is a hip that can be reduced concentrically and congruently. The hip must also have near-normal ROM. Redirectional osteotomies include single innominate (Salter), triple innominate (Steel), and peri-acetabular (Ganz) (
Figure 7). Reshaping osteotomies include Pemberton and Dega (
(b) Salvage procedures are typically indicated in adolescents with severe dysplasia in whom acetabular deficiency precludes the use of a redirectional osteotomy. In salvage procedures, weight-bearing coverage is increased by using the joint capsule as an interposition between the femoral head and bone above it. The intent of these osteotomies is to reduce point loading at the edge of the acetabulum. These osteotomies rely on fibrocartilaginous metaplasia of the interposed joint capsule to provide an increased articulating surface. Salvage osteotomies include Chiari (
Figure 9) and shelf osteotomies.
[Table 2. Obstacles to a Concentric Reduction of the Hip in DDH]
[Table 3. Advantages and Disadvantages of Anterior Versus Medial or Anteromedial Approaches]
[Table 4. Pelvic Osteotomies for the Treatment of DDH]
[Figure 7. Redirectional pelvic osteotomy options. A, Single innominate (Salter); B, triple innominate; C, periacetabular.]
[Figure 8. Reshaping pelvic osteotomy options. A, Pemberton; B, Dega.]
[Figure 9. Salvage pelvic osteotomy—Chiari medial displacement osteotomy.]
II. Legg-Calve-Perthes Disease
1. Definition—Legg-Calve-Perthes disease (LCPD) is an idiopathic osteonecrosis of the capital femoral epiphysis in children.
a. LCPD affects 1 in 1,200 children.
b. The disease more commonly affects boys than girls (4:1 to 5:1).
c. The hips are involved bilaterally in 10% to 12% of cases.
d. LCPD is more commonly diagnosed in urban than rural communities.
e. There appears to be a predilection in certain populations, with a higher incidence in Asians, Eskimos, and central Europeans. Incidence is lower in native Australians, Native Americans, Polynesians, and African-Americans.
a. The exact etiology of LCPD is unknown.
b. Historically, the cause was thought to be inflammatory or infectious in nature, with transient synovitis as a possible precursor. Trauma was also thought to be causative at one time.
c. Current theories propose that a disruption of the vascularity of the capital femoral epiphysis occurs, resulting in necrosis and subsequent revascularization.
i. Deficient vascularity may be due to interruption of the blood supply to the femoral head.
ii. The vascularity of the capital femoral epiphysis may also be threatened by thrombophilia and/or various coagulopathies (protein C and S deficiency, activated protein C resistance).
(a) Thrombophilia has been reported to be a possible causative factor in 50% of children with LCPD.
(b) As many as 75% of patients with LCPD will have a coagulopathy.
2. Risk factors
a. Family history is positive in 1.6% to 20% of cases.
b. LCPD is associated with attention-deficit hyperactivity disorder (33%).
c. Patients are commonly skeletally immature, with bone age delayed in 89% of cases.
3. Pathology—The capital epiphysis and physis are abnormal histologically, with disorganized cartilaginous areas of hypercellularity and fibrillation.
1. Clinical presentation
a. LCPD occurs most commonly in children from age 4 to 8 years (range, 2 years to late teens).
b. Onset is insidious, and children with LCPD will commonly have a limp and pain in the groin, hip, thigh, or knee regions.
c. Occasionally, children with LCPD will have a history of recent or remote viral illness.
2. Physical examination
a. Examination of the child with LCPD may reveal an abnormal gait (antalgic and/or Trendelenburg).
b. ROM testing will often reveal decreased abduction and internal rotation. Hip flexion contractures are seen rarely.
c. Limb-length inequality, if present, is mild secondary to femoral head collapse. The presence of hip contractures may make the limb-length inequality appear greater than it actually is.
3. Diagnostic tests
a. Plain radiographs
i. Standard AP and frog-leg lateral views of the pelvis are critical in making the initial diagnosis and assessing the subsequent clinical course.
ii. LCPD typically proceeds through four radiographic stages.
(a) Initial stage—Early radiographic findings are a sclerotic, smaller proximal femoral ossific nucleus (due to failure of the epiphysis to increase in size) and widened medial clear space (distance between teardrop and femoral head).
(b) Fragmentation stage—Segmental collapse (resorption) of the capital femoral epiphysis follows, with increased density of the epiphysis.
(c) Reossification or reparative stage—Necrotic bone is resorbed with subsequent reossification of the capital femoral epiphysis.
(d) Remodeling stage—Remodeling begins when the capital femoral epiphysis is completely reossified.
b. Other imaging studies—Other imaging modalities, including bone scans, MRI, and arthrography, are not routinely necessary.
i. Bone scan—Can confirm a suspected diagnosis of LCPD (and the extent of femoral head involvement). Decreased uptake in the capital femoral epiphysis ("cold" lesion, suggesting decreased blood flow) is one of the earliest findings in LCPD and can predate changes on plain radiographs.
ii. MRI—Can also aid in the early diagnosis of LCPD, revealing areas of decreased signal intensity in the capital femoral epiphysis and alterations in the physis.
iii. Arthrography (especially dynamic)—A useful modality to assess coverage and containment of the femoral head. Arthrography is often used at the time of surgery and confirms the degree of correction needed for femoral and/or pelvic osteotomies.
1. The Herring (lateral pillar) classification is based on the height of the lateral pillar of the capital epiphysis on AP view of the pelvis (
a. Group A—No involvement of the lateral pillar, with no density changes and no loss of height of the lateral pillar.
b. Group B—More than 50% of the lateral pillar height is maintained.
c. Group C—Less than 50% of the lateral pillar height is maintained.
d. B/C border group—This group has been added more recently to increase the consistency of readings and to increase prognostic accuracy of the lateral pillar classification. In this group, the lateral pillar is narrow (2 to 3 mm wide) or poorly ossified, or exactly 50% of lateral pillar height is maintained.
e. The Herring classification is the most reliable classification scheme and is related to prognosis. Its limitation is that the final classification cannot be determined at the time of initial presentation.
2. The Catterall classification is based on the amount of epiphyseal involvement (
Figure 11). Although commonly used in the past, it has more recently been criticized for its poor interobserver reliability.
a. Group I—Involvement is limited to the anterior part of the capital epiphysis.
b. Group II—The anterior and central parts of the capital epiphysis are involved.
c. Group III—Most of the capital epiphysis is involved, with sparing of the medial and lateral parts of the epiphysis.
[Figure 10. Herring lateral pillar classification for LCPD. Group A hips have no evident involvement of the lateral pillar (lateral third of epiphysis; shaded area) and the lateral pillar height is normal. Group B hips have loss of lateral pillar height < 50%. Type C hips have lost > 50% of the lateral pillar height.]
d. Group IV—The entire epiphysis is involved.
e. Catterall also described four at-risk signs, which indicate a more severe disease course.
i. Gage sign (radiolucency in the shape of a V in the lateral portion of the epiphysis)
ii. Calcification lateral to the epiphysis
iii. Lateral subluxation of the femoral head
iv. A horizontal physis
a. Treatment is controversial.
b. Most patients (60%) with LCPD will not require treatment.
c. Patients with a good prognosis will not usually require treatment.
[Figure 11. Catterall classification for LCPD.]
i. Good prognosis is expected for patients with Catterall type I and Herring group A disease.
ii. Young age (<8 years) at onset of disease is also a good prognostic factor.
d. Patients with a poor prognosis will usually need treatment.
i. Prognosis is poor for patients with Catterall types III and IV and Herring groups B and C disease.
ii. Children older than 8 years at onset of disease have a guarded prognosis.
e. All patients require careful periodic clinical and radiographic assessments.
a. For patients with a good prognosis, including younger patients (<8 years of age at initial presentation) with containable hips (ie, femoral heads that are seated well in the acetabulum), symptomatic and supportive measures are usually adequate.
i. Preservation of ROM is essential.
ii. Protected weight bearing has also been recommended, especially before the reossification stage.
b. For patients with a poor prognosis or for patients with progressive deformity and/or progressive loss of motion, particularly abduction, the basis for treatment is containment.
i. The aim of containment is to ensure that the femoral head remains well seated in the acetabulum.
ii. In the long term, the goal of treatment is to prevent deformity and thereby prevent degenerative joint disease.
3. Containment treatment
i. Containment may be achieved by nonsurgical means, specifically by casting or bracing in an abducted and internally rotated position.
ii. Petrie casts and a variety of abduction orthoses have been used.
iii. The long-term benefit of casting and bracing has been called into question.
i. Several surgical methods (described below) have been recommended to preserve containment. The results of these methods are comparable.
ii. For these methods to be effective, the hips must be "containable," ie, relatively full ROM with congruency between the femoral head and acetabulum.
iii. Surgical containment may be approached from the femoral side, the acetabular side, or both sides of the hip joint, according to surgeon preference.
(a) On the femoral side, a proximal femoral varus osteotomy is an option.
(b) On the acetabular side, a pelvic osteotomy (Salter, triple innominate, Dega, or Pemberton) may be performed.
(c) A shelf arthroplasty can also be performed to prevent subluxation and lateral overgrowth of the capital epiphysis.
(d) An arthrodiastasis (hip distraction) for 4 to 5 months has also been advocated in some centers, although this is more complicated and cumbersome for the patient.
iv. Once the hip is no longer containable, or for patients presenting late with deformity, treatment strategy focuses on salvage procedures, and the goals of treatment are relief of symptoms and restoration of stability.
(a) These patients are at risk for hinge abduction, ie, lateral extrusion of the femoral head resulting in the femoral head impinging on the edge of the acetabulum with abduction.
(b) An abduction-extension proximal femoral osteotomy should be considered.
(c) Pelvic osteotomy procedures such as a Chiari osteotomy and shelf arthroplasty may be beneficial.
a. The wide range of complications includes femoral head deformity, premature physeal arrest patterns, osteochondritis dissecans, labral injury, and late arthritis.
b. The most important prognostic factors are the shape of the femoral head and its congruency at skeletal maturity and patient age at onset of disease. Stulberg correlated worse long-term outcomes to greater deformities in the femoral head at maturity.
c. Long-term follow-up studies indicate that most patients with LCPD do well until the fifth or sixth decade of life, at which time degenerative changes in the hip joint are common.
III. Slipped Capital Femoral Epiphysis
a. Slipped capital femoral epiphysis (SCFE) is a disorder of the hip in which the femoral neck displaces anteriorly and superiorly relative to the femoral epiphysis.
b. Displacement occurs through the proximal femoral physis.
a. SCFE is the most common disorder of the hip in adolescents. It occurs in 1 to 61 per 100,000 children annually.
b. Males are more commonly affected than females (2:1). The cumulative risk in males is 1 per 1,000 to 2,000; in females, 1 per 2,000 to 3,000.
c. The left hip is more commonly involved.
d. Unilateral involvement at time of presentation is more common (80%). Ultimately, the hips are involved bilaterally in 10% to 60% of cases.
e. SCFE occurs more commonly in Polynesians and African-Americans.
a. The precise etiology is unknown.
b. In general, SCFE is thought to be the result of mechanical insufficiency of the proximal femoral physis to resist load. This can occur because of either physiologic loads across an abnormally weak physis or abnormally high loads across a normal physis.
i. Conditions that weaken the physis include endocrinopathies such as hypothyroidism, panhypopituitarism, growth hormone abnormalities, hypogonadism, and hyper- or hypoparathyroidism; systemic diseases such as renal osteodystropy; and a history of previous radiation therapy to the femoral head region.
Figure 12. Radiographs of a 10-year old girl with a stable SCFE on the left. A, AP view demonstrates that the Klein line intersects the epiphysis bilaterally. B, Frog-leg lateral view of the same patient, however, more clearly demonstrates the left SCFE.]
ii. Several mechanical factors that can increase the load across the physis are associated with SCFE.
(a) SCFE occurs more commonly in overweight children, with 50% of patients at or above the 90th percentile for weight and 70% of patients above the 80th percentile for weight.
(b) Decreased femoral anteversion and decreased femoral neck-shaft angle are also associated with SCFE.
2. Pathology—The physis is abnormally widened with irregular organization. The slip occurs through the proliferative and hypertropic zones of the physis.
1. Clinical presentation
a. SCFE occurs most commonly in children 10 to 16 years of age.
i. In boys, age of presentation is between 12 and 16 years (average, 13.5 years).
ii. In girls, age of presentation is between 10 and 14 years (average, 11.5 years).
b. Children with SCFE commonly have a limp and pain in the groin, hip, thigh, or knee region.
i. Pain is localized to the distal thigh and/or knee region in 23% to 46% of cases.
ii. Symptoms are usually present for weeks to several months before a diagnosis is made.
2. Physical examination
a. Common physical findings include an abnormal gait (antalgic and/or Trendelenburg) and decreased ROM (in particular, decreased hip flexion and decreased internal rotation).
b. ROM testing may also reveal obligate external rotation, ie, external rotation of the hip as the hip is brought into flexion.
c. The foot and knee progression angles are usually externally rotated.
3. Diagnostic tests
a. Plain radiographs—Standard AP and frog-leg lateral views of the pelvis are recommended.
i. The Klein line, a line tangential to the superior border of the femoral neck on the AP view, intersects the proximal femoral epiphysis in a normal hip. The Klein line may fail to intersect the proximal femoral epiphysis in a hip involved with SCFE or will be asymmetric in the two hips (Figure 12, A).
ii. Lateral radiographs are more sensitive in detecting a SCFE (Figure 12, B).
iii. Radiographic findings also include the metaphyseal blanch sign—superimposition of the posteriorly displaced epiphysis on the femoral neck (seen on AP view).
b. Other imaging studies
i. MRI may be useful in diagnosing "pre-slip" hips. An abnormally widened physis with surrounding edematous changes on MRI are suggestive of pre-slip hips.
ii. Although usually not necessary for the diagnosis of SCFE, MRI may be helpful in the evaluation of osteonecrosis afterward.
iii. CT can be useful in characterizing the proximal femoral deformity—especially during preoperative planning for reconstructive procedures—although it is generally not needed.
1. The Loder classification is the most widely used classification and is based on SCFE stability (
a. The SCFE is stable if the patient is able to weight bear on the involved extremity (with or without crutches).
b. The SCFE is unstable if the patient is unable to weight bear on the involved extremity.
i. The value of the Loder classification is its superior ability to predict osteonecrosis.
ii. Based on a single study, the risk of osteonecrosis in unstable hips was reported as 47%; in stable hips, 0%.
2. The traditional classification is temporal and based on duration of symptoms but has largely been replaced by the Loder classification because of the superior prognostic value of the latter.
a. The SCFE is chronic when symptoms have been present for >3 weeks.
b. The SCFE is acute when symptoms have been present for <3 weeks.
c. The SCFE is acute-on-chronic when there is an acute exacerbation of symptoms following a prodrome of at least 3 weeks.
3. Radiographic classifications
a. Depending on the amount of slip (percentage of epiphyseal displacement relative to metaphyseal width of femoral neck [on AP or lateral radiographs]), the SCFE is mild (0% to 33%), moderate (33% to 50%), or severe (>50%).
b. Depending on the difference in Southwick angle between the involved and uninvolved sides, the SCFE is mild (<30° difference), moderate (30° to 50°), or severe (>50°). The Southwick angle, or the head-shaft angle, is the angle formed by the proximal femoral physis and femoral shaft on lateral radiographs.
1. Nonsurgical—Nonsurgical management, including spica casting, is no longer recommended. Complication rates were high and included chondrolysis, further slip, and osteonecrosis.
In situ screw fixation is the preferred initial treatment of SCFE (
Forceful manipulation is never indicated because it is associated with an increased risk of complications, including osteonecrosis.
[Table 5. Classification of SCFE]
Serendipitous or gentle reduction does not appear to negatively affect patient outcomes.
For a stable SCFE, a single-screw construct is usually adequate. For an unstable SCFE, use of two screws should be considered, though this entails an increased risk of implant penetration.
Technical points—Cannulated screw systems are convenient. There are two reasons for starting the screw(s) on the anterior femoral neck: First, because the femoral head has slipped posteriorly, an anterior starting point allows the screw to be targeted to the center position of the femoral head and perpendicular to the physis on both AP and lateral views. Second, a lateral entry point (particularly at or below the lesser trochanter) increases the risk of postoperative fracture. Because of the "blind spot," screws in the center-center position must be at least 5 mm from subchondral bone in all views, and screws in any other position must be at least 10 mm from subchondral bone.
Bone peg epiphysiodesis is an alternative treatment option. Given its increased complexity and high complication rates (including blood loss), bone peg epiphysiodesis has fallen out of favor in most centers.
For severe deformities, a proximal femoral osteotomy at the subcapital, femoral neck, intertrochanteric, or subtrochanteric level can be performed. Osteotomy at the subcapital and femoral neck levels can provide the most correction but in general should be avoided because the complication rates are highest among all osteotomies.
Indications for prophylactic fixation of the contralateral hip include age younger than 10 years and associated at-risk conditions
[Figure 13. Postoperative AP (A) and frog-leg lateral (B) views after in situ screw fixation of a stable SCFE.]
such as endocrinopathies, renal osteodystrophy, and a previous history of radiation therapy.
i. Osteonecrosis (ON)—The most accurate predictor of ON is the stability of the slip, with up to 47% risk of ON for unstable slips. However, hardware placement in the posterior and superior femoral neck can disrupt the interosseous blood supply and also cause osteonecrosis.
ii. Chondrolysis—Chondrolysis is due to unrecognized implant penetration of the articular surface. If penetration is recognized at the time of surgery and corrected, chondrolysis does not occur. Chondrolysis was a common complication of spica casting in children with SCFE before casting fell out of favor.
iii. Slip progression—Slip progression occurs in 1% to 2% of cases following in situ single-screw fixation.
iv. Fracture—The risk of fracture is increased with entry sites through the lateral cortex and those at or distal to the lesser trochanter.
3. Rehabilitation—Weight bearing is usually protected postoperatively.
a. With an unstable SCFE, 4 to 6 weeks of non-weight-bearing precautions is recommended.
b. With a stable SCFE, recommendations for weight bearing are variable but usually involve a brief period of partial weight-bearing precautions.
IV. Coxa Vara
1. Definition—Coxa vara is defined as an abnormally low femoral neck-shaft angle (<120°).
i. Congenital—Congenital coxa vara is characterized by a primary cartilaginous defect in the femoral neck. It commonly is associated with congenital short femur, congenital bowed femur, and proximal femoral focal deficiency (PFFD; also known as partial longitudinal deficiency of the femur).
ii. Acquired—Coxa vara can be secondary to numerous conditions including trauma, infection, pathologic bone disorders (eg, osteopetrosis), SCFE, LCPD, and skeletal dysplasias.
iii. Developmental—Developmental coxa vara occurs in early childhood, with classic radiographic changes (including the inverted Y sign) and no other skeletal manifestations. The remainder of the coxa vara section focuses on developmental coxa vara.
a. Coxa vara occurs in 1 in 25,000 live births worldwide.
b. Boys and girls are affected equally.
c. The right and left sides are affected equally.
d. Bilateral involvement occurs in 30% to 50% of cases.
e. There is no major variation in incidence by race.
1. The exact cause of coxa vara remains unknown.
2. There appears to be a genetic predisposition, with an autosomal dominant pattern and incomplete penetrance.
3. The most widely accepted theory attributes the deformity in the proximal femur to a primary defect in endochondral ossification in the medial part of the femoral neck.
a. The eventual dystrophic bone along the medial inferior aspect of the femoral neck fatigues with weight bearing, resulting in a progressive varus deformity.
b. The vertical orientation of the proximal femoral physis converts normal compressive forces across the physis to a greater shear force. In addition, compressive forces across the medial femoral neck are increased.
1. Clinical presentation
a. The child usually presents after walking is started and before 6 years of age.
b. Pain is uncommon.
c. An apparent limb shortening or painless limp may be present in unilateral cases. A waddling gait is more characteristic in bilateral cases.
2. Physical examination
a. Physical findings include a prominent greater trochanter, which also may be more proximal relative to the contralateral side.
b. With unilateral involvement, limb-length inequality (usually minor, <3 cm) may be present.
c. Abductor muscle weakness is common. Consequently, Trendelenburg gait may be present and the Trendelenburg sign may be positive.
d. ROM testing may demonstrate decrease in abduction and internal rotation.
e. With bilateral involvement, lumbar lordosis may be increased.
3. Plain radiographs—AP and frog-leg lateral views of the pelvis are recommended. Radiographic findings include:
a. A decreased femoral neck-shaft angle
b. The inverted Y sign (resulting from a triangular metaphyseal fragment in the inferior femoral neck), which is pathognomic (
c. Vertical orientation of the physis, a shortened femoral neck, and decreased femoral anteversion
[Figure 14. The inverted Y sign (arrows), a triangular metaphyseal fragment in the inferior femoral neck, is pathognomic for coxa vara.]
Figure 15. The Hilgenreiner-epiphyseal angle is formed by a line through the physis and the Hilgenreiner line. The normal angle is about 25°.]
d. Abnormal Hilgenreiner-epiphyseal angle, the angle formed by the physis and Hilgenreiner's line. This angle correlates with the risk of disease progression (Figure 15).
D. Treatment—Treatment recommendations are based on the severity of the Hilgenreiner-epiphyseal angle and presence of symptoms.
a. Asymptomatic patients with a Hilgenreiner-epiphyseal angle <45° should be observed.
b. Asymptomatic patients with a Hilgenreiner-epiphyseal angle between 45° and 59° can also be observed. Serial radiographs are critical to assess for progression.
i. Patients with a Trendelenburg gait and/or fatigue pain in the hip abductors and a Hilgenreiner angle between 45° and 59° or those with evidence of progression
ii. Patients with a Hilgenreiner-epiphyseal angle >60°
iii. Patients with a progressive decrease in the femoral neck-shaft angle to 90° or 100° or less
i. The standard procedure is a proximal femoral valgus derotational osteotomy
(a) The osteotomy can occur at the intertrochanteric or subtrochanteric level as described by Langenskiold (intertrochanteric), Pauwel (Y-shaped intertrochanteric), and Borden (subtrochanteric).
(b) Osteotomy at the femoral neck level should be avoided because higher morbidity rates and poorer clinical results have been reported.
ii. The ultimate goal of surgery is to provide valgus overcorrection of the femoral neck-shaft angle (Hilgenreiner-epiphyseal angle <38°).
iii. Adductor tenotomy is frequently necessary.
a. Varus deformity recurs following valgus osteotomy in up to 50% of cases. The risk of recurrence may be decreased by valgus overcorrection.
b. Premature closure of the proximal femoral physis has been reported in up to 89% of cases.
i. Premature closure is usually noted within the first 12-24 months after surgery.
ii. Premature closure may lead to limb-length inequality and/or trochanteric overgrowth.
4. Rehabilitation—Spica cast immobilization is recommended for 6 to 8 weeks after surgery.
Beals RK: Coxa vara in childhood: Evaluation and management. J Am Acad Orthop Surg 1998;6:93-99.
Carney BT, Weinstein SL, Noble J: Long-term follow-up with slipped capital femoral epiphysis. J Bone Joint Surg Am 1991; 73:667-674.
Carroll K, Coleman S, Stevens PM: Coxa vara: Surgical outcomes of valgus osteotomies. J Pediatr Orthop 1997;17:220-224.
Gillingham BL, Sanchez AA, Wenger DR: Pelvic osteotomies for the treatment of hip dysplasia in children and young adults. J Am Acad Orthop Surg 1999;7:325-337.
Herring JA, Hui K, Browne R: Legg-Calve-Perthes disease: Classification of radiographs with use of the modified lateral pillar and Stulberg classifications. J Bone Joint Surg Am 2004;86:2103-2120.
Herring JA, Hui K, Browne R: Legg-Calve-Perthes disease: Prospective multicenter study of the effect of the treatment on outcome. J Bone Joint Surg Am 2004;86:2121-2133.
Kocher MS, Bishop JA, Weed B, et al: Delay in diagnosis of slipped capital femoral epiphysis. Pediatrics 2004;113:e322-e325.
Loder RT: Unstable slipped capital femoral epiphysis. J Pediatr Orthop 2001;21:694-699.
McAndrew MP, Weinstein SL: A long-term follow-up of Legg-Calve-Perthes disease. J Bone Joint Surg Am 1984;66: 860-869.
Schultz WR, Weinstein JN, Weinstein SL, Smith BG: Prophylactic pinning of the contralateral hip in slipped capital femoral epiphysis: Evaluation of long-term outcome for the contralateral hip with use of decision analysis. J Bone Joint Surg Am 2002;84:1305-1314.
Skaggs DL, Tolo VT: Legg-Calve-Perthes disease. J Am Acad Orthop Surg 1996;4:9-16.
Staheli LT: Surgical management of acetabular dysplasia. Clin Orthop Relat Res 1991;264:111-121.
Vitale MG, Skaggs DL: Developmental dysplasia of the hip from six months to four years of age. J Am Acad Orthop Surg 2001;9:401-411.
Weinstein JN, Kuo KN, Millar EA: Congenital coxa vara: A retrospective review. J Pediatr Orthop 1984;4:70-77.
Weinstein SL, Mubarak SJ, Wenger DR: Developmental hip dysplasia and dislocation: Part I. J Bone Joint Surg Am 2003; 85:1824-1832.
Weinstein SL, Mubarak SJ, Wenger DR: Developmental hip dysplasia and dislocation: Part II. J Bone Joint Surg Am 2003; 85:2024-2035.
Wientroub S, Grill F: Ultrasonography in developmental dysplasia of the hip. J Bone Joint Surg Am 2000;82:1004-1018.
Top Testing Facts
Developmental Dysplasia of the Hip
1. Because the ossific nucleus of the femoral head does not appear until 4 to 6 months of age, ultrasound is a better test than plain radiographs in the first 4 to 6 months of life.
2. Routine ultrasound screening should be performed for infants with risk factors for the condition.
3. Because of the poor specificity of ultrasonography in children younger than 1 month, hip ultrasonography should be deferred until after 1 month of life.
4. Pavlik harness treatment should be discontinued if the dislocated hip does not relocate within 3 to 4 weeks, to avoid Pavlik harness disease.
5. Excessive hip flexion in the Pavlik harness results in an increased risk of femoral nerve palsy.
6. Excessive hip abduction in the Pavlik harness results in an increased risk of osteonecrosis of the femoral head.
7. The Galeazzi (or Allis) test is positive in unilateral dislocation but not bilateral dislocations.
8. Because asymmetric skin folds are common in children with normal hips, children with such asymmetry have a high rate of false positives.
9. Hip abduction does not become limited in DDH until approximately 6 months of age.
1. The most important prognostic factors are the shape of the femoral head and its congruency at skeletal maturity and patient age at onset of disease. Stulberg correlated worse long-term outcomes to greater deformities in the femoral head at maturity.
2. The Herring lateral pillar classification, based on preservation of the height and integrity of the lateral pillar of the femoral head, is the most reliable classification scheme and is related to prognosis. Its limitation is that a final staging cannot be accurately determined at the time of presentation.
3. Most hips (60%) will not require treatment other than symptomatic and supportive measures.
4. Most patients with LCPD will do well until the 5th to 6th decades of life.
5. Treatment continues to be controversial. For surgical containment methods, the results of proximal femoral osteotomy, pelvic osteotomy, and shelf arthroplasty procedures appear comparable.
Slipped Capital Femoral Epiphysis
1. The frog-leg lateral radiograph is the most sensitive for detecting SCFE.
2. During in situ screw fixation, the starting point should be positioned anteriorly because the femoral epiphysis is posterior relative to the femoral neck.
3. Because of an increased risk of fracture, a lateral entry point should be avoided, especially when at or distal to the lesser trochanter.
4. Because of the "blind spot," screws in the center-center position must be at least 5 mm from subchondral bone on all views—at least 10 mm when the screw is not in the center-center position.
5. The most accurate predictor of osteonecrosis is the stability of the hip at presentation; up to 47% risk of osteonecrosis is associated with unstable slips.
6. Chondrolysis is a consequence of unrecognized permanent screw penetration. If screw penetration is noted at time of surgery and corrected, there is no increased risk of chondrolysis. Before casting fell out of favor, chondrolysis was commonly seen after cast treatment of SCFE as well.
1. The inverted Y sign on radiographs is pathognomic for the diagnosis of coxa vara.
2. The Hilgenreiner-epiphyseal angle is prognostic and critical in treatment decision making. Surgery is indicated for an angle >60° and observation for an angle <45°. Those with angles between 45° and 59° require observation for potential progression.
3. A successful outcome following surgery is dependent on valgus overcorrection of the proximal femoral deformity.