Diaphyseal Fractures: Although multiple fractures at different stages of healing are strongly suggestive of child abuse, the most common manifestation is an isolated diaphyseal fracture, which occurs four times as often as classic metaphyseal fractures. The humerus, femur, and tibia are the most frequently fractured long bones; the radius and ulna are the most infrequently fractured. With the exception of supracondylar fractures, all fractures of the midshaft of the humerus in children younger than 3 years are strongly suggestive of abuse.

Metaphyseal Fractures: Although less common than diaphyseal fractures, metaphyseal fractures are more specific for child abuse. Metaphyseal fractures most commonly affect the tibia, femur, and proximal end of the humerus. Corner fractures and bucket-handle fractures, which are probably the same fracture viewed in two different projections, result from violent shaking or forceful pulling or twisting of an infant’s limb. The diagnosis is made by careful evaluation of high-quality plain radiographs. The tight adherence of the periosteum at the metaphysis precludes an active periosteal response, making these fractures difficult to diagnose even in the healing stages. In addition, after healing, these fractures may not be visible because of rapid bone remodeling. Bone scans, although sometimes helpful, are difficult to interpret because of the normally increased radionuclide uptake in the metaphyseal area.

Rib Fractures: Rib fractures are present in 5 to 27% of cases of child abuse; 90% of such fractures occur in children younger than 2 years. The young pediatric rib cage is compliant, and considerable force is required to break a rib. Accordingly, rib fractures are seldom seen in unintentional injury and are almost never seen after cardiopulmonary resuscitation.³⁰ When present, postresuscitation rib fractures are anterior and also may be multiple.³⁰ In a child younger than 3 years, the positive predictive value of a rib fracture as an indicator of intentional trauma approaches 100%.³¹

With nonaccidental trauma, posterior rib fractures are most common and result from maximal mechanical stress as the rib is levered over the transverse process of the vertebral body when infants are grasped and shaken. The ribs fail mechanically at the head or neck. Abuse-related rib fractures tend to be multiple and symmetrical and may be difficult to diagnose acutely on standard radiographs. Repeated radiographic examination at 7 to 10 days after the injury is advised if findings on the initial films are normal but the level of clinical suspicion is high. Radiographic findings include callus formation or widening of the rib neck as a result of apposition of new bone subperiosteally. Bone scans can be helpful in detection of fractures in the acute setting.

Skull Fractures: Skull fractures are the second most frequent injury in abuse and occur more commonly in abuse cases than from unintentional trauma.²⁶ Eighty percent occur in infants younger than 1 year, and although complex skull fractures are more suspicious for abuse, linear skull fractures are the most common type. Standard skull radiographs may not be adequate for diagnosis and are reported to miss more than 25% of head injuries. Children who meet high-risk criteria (the presence of rib fractures, multiple fractures or facial injury, or age younger than 6 months) should undergo CT or MRI for assessment of occult head injury.³²

Periosteal New Bone Formation: Periosteal new bone formation, which is one of the most common findings in cases of abuse, reflects separation of the periosteum from the bone and may be the only manifestation of orthopedic trauma related to child abuse. It may be present with or without a fracture. It results from shaking, from acceleration-deceleration forces applied to an unsupported limb, or from forceful gripping.

Transient Synovitis: Transient synovitis is the most common cause of hip pain in childhood.⁴¹ It is a self-limited condition caused by a nonpyogenic inflammatory response of the synovium. Although it has been reported in children as young as 3 months and occasionally occurs in adults,⁴² its peak incidence is between 3 and 6 years of age. Transient synovitis of the hip affects boys more commonly than girls and has a slight predilection for the right side. Less than 5% of cases are bilateral. Approximately half of the patients are seen acutely, usually within the first 3 days of the onset of symptoms, whereas the other half have a more insidious course, with symptoms occurring for weeks to months before medical evaluation.⁴³ Although transient synovitis most commonly affects the hip, it also can affect the knee.

The etiology of transient synovitis is unknown. Current theories imply an association with active or recent infection, trauma, or allergic hypersensitivity. At least half of the children with transient synovitis have or recently have had an upper respiratory illness.⁴⁴ One study demonstrated a fourfold rise in viral titers in 45% of patients and elevated serum interferon levels, consistent with a concurrent viral infection, in 43% of patients with transient synovitis.⁴⁵ Trauma commonly is associated with transient synovitis, but no clear relationship has been demonstrated. Similarly, a causal association with an allergic reaction to an infectious agent has been suggested but not proved.

It is estimated that transient synovitis of the hip may occur in up to 3% of children. Hip or groin pain is the most common initial finding, but referred pain to the medial aspect of the thigh or knee is found in 10 to 30% of patients. Affected patients either walk with a limp or, with severe pain, refuse to walk at all. The leg is held in flexion with slight abduction and external rotation. On examination, passive movement usually is pain free; however, pain and a slightly decreased range of motion may be noted with extreme internal rotation or abduction. Although most children with transient synovitis tend to otherwise be well, some will have a low-grade fever and malaise.

The diagnosis of transient synovitis is one of exclusion and relies on the history and physical examination in combination with limited laboratory testing and AP and frog-leg lateral radiographs of the pelvis. Laboratory tests are used to help distinguish children with transient synovitis from those with septic arthritis. In transient synovitis, laboratory values may be normal or may reveal mild elevations in the white blood cell count and erythrocyte sedimentation rate (ESR), both consistent with a nonspecific inflammatory process. Multiple studies have attempted to define criteria to help differentiate transient synovitis from septic arthritis. Clinical decision rules have included the presence or absence of fever, the ability to bear weight, white blood cell counts, the inflammatory markers ESR and C-reactive protein (CRP), side-to-side differences in the width of the joint space on radiographs, and previous visits to a health care provider for related symptoms. Kocher and colleagues found four independent multivariant predictors of septic arthritis: fever, inability to bear weight, ESR of 40 mm/hr or higher, and serum white blood cell count higher than 12,000 cells/µL. Patients with three of the four predictors had a 93% chance of having septic arthritis, and those with all four had a 99% likelihood.⁴⁶ A subsequent study to validate these findings showed that patients with none of the predictors had a 2% probability of septic arthritis; with one of four predictors, a 9.5% probability; with two of four predictors, a 35% probability; with three of four predictors, a 73% probability; and with all four predictors, a 93% probability of septic arthritis.⁴⁷ When this algorithm was tested at another institution, however, the presence of all four Kocher criteria predicted septic arthritis only 59% of the time.⁴⁸ In general, in children with transient synovitis and septic arthritis, the overlap of laboratory values and historical features is too great to provide a foolproof diagnostic algorithm.

Although radiographs of the hip and pelvis tend to be normal in appearance with transient synovitis, they are helpful to exclude other diseases. Radiographic findings consistent with transient synovitis include medial joint space widening, accentuated pericapsular shadow, and Waldenström’s sign, which is lateral displacement of the femoral epiphysis with surface flattening secondary to effusion. However, these findings also are apparent in Legg-Calvé-Perthes disease and, if present, mandate close follow-up or further investigation with MRI. If joint aspiration is necessary to clarify the diagnosis, ultrasonography can be used to guide hip joint aspiration. Effusions are present in 60 to 70% of cases of transient synovitis; however, they also are present with septic arthritis, osteomyelitis, acute slipped capital femoral epiphysis (SCFE), Legg-Calvé-Perthes disease, rheumatoid and infectious arthritis, malignant disease, and osteoid osteoma. Accordingly, the presence of an effusion on ultrasonography cannot be used to distinguish transient synovitis from other causes of hip pain.⁴⁹ Nuclear scintigraphy in patients with transient synovitis is helpful in differentiation of transient synovitis from osteomyelitis (with or without septic arthritis), Legg-Calvé-Perthes disease, and SCFE. Scintigraphy may demonstrate differential uptake in the capital femoral epiphysis, with decreased uptake early in the course of the disease and increased uptake later on. The early changes also are seen with very early Legg-Calvé-Perthes disease and suggest that some children with transient synovitis experience ischemia of the capital femoral epiphyses. This ischemia may be caused by an intracapsular effusion that tamponades vessels along the femoral neck. These ischemic findings have not yet been shown to have clinical significance.

Most cases of transient synovitis can be managed at home with close follow-up by the child’s primary care provider. In general, these children have a mild limp, and treatment with nonsteroidal anti-inflammatory medication often results in clinical improvement with relief of symptoms. A complete blood count and determination of ESR should be performed, and radiographs should be obtained. Patients with severe symptoms, fever, and elevated ESR should be evaluated for septic arthritis. If there is any doubt about the diagnosis, immediate orthopedic consultation is necessary.

Treatment of transient synovitis is twofold: (1) rest of the affected joint achieved with non-weightbearing or, in cases associated with extreme pain, bed rest and (2) reduction of the inflammatory process with anti-inflammatory medications. Temperatures should be monitored closely, and any fever should be reported to the physician. Children are allowed a gradual return to activity as the pain subsides, and full unrestricted activity is permitted when the hip is completely pain free with no evidence of a limp. Although many children with transient synovitis have an effusion, therapeutic aspiration of the joint is not routinely performed, and there is no strong evidence that aspiration shortens the clinical course or prevents osteonecrosis. Repeated examination is recommended for all children within 12 to 24 hours and then again after 10 to 14 days if the symptoms have not resolved.

The prognosis for children with transient synovitis is excellent. Up to 75% of patients have complete resolution of pain within 2 weeks and 88% within 4 weeks. The remainder may have less intense but persistent pain for up to 8 weeks. Relapse is possible, although infrequent, and usually occurs within 6 months. In general, long-term sequelae of transient synovitis are infrequent; these may include asymptomatic coxa magna (enlargement and deformity of the femoral head and neck caused by hypertrophy of cartilage secondary to inflammation) and mild degenerative cystic changes of the femoral neck. These changes persist for years and may be accompanied by radiographic degenerative change, but they do not tend to cause long-term functional disability. In addition, a small number (2%) of cases of transient synovitis follow a clinical and radiographic course consistent with Legg-Calvé-Perthes disease. Whether this is due to ischemia of the capital femoral epiphyses during the early stages of synovitis or reflects an initial misdiagnosis is unclear. It is recommended that children with persistent symptoms undergo ultrasonography to evaluate for the presence of an effusion. Persistent joint effusion beyond 4 to 6 weeks may be associated with the subsequent development of Legg-Calvé-Perthes disease.⁵⁰ Some experts recommend routine 6-month follow-up radiographs for all children with transient synovitis,⁴³ whereas others reserve radiographs for children who remain symptomatic.⁵¹

Acute Septic Arthritis: Septic arthritis refers to microbial invasion and infection of the joint space. Bacterial pathogens are common in patients with acute septic arthritis, whereas fungal and mycobacterial pathogens tend to be associated with a more indolent septic arthritis. Acute septic arthritis occurs in all age groups but is more common in children; 70% of cases occur in children younger than 4 years, and the peak incidence is between 6 and 24 months. Boys are affected twice as frequently as girls. Predisposing factors include preceding viral infection, trauma, immunodeficiency, hemoglobinopathy, hemophilia with recurrent hemarthroses, diabetes, injection drug abuse, rheumatoid arthritis, and intra-articular injections or operations. In 75% of cases of septic arthritis, the pathologic process involves the joints of the lower extremity, with the knee being most commonly and the hip second most commonly involved. Other affected joints, in order of involvement, include the ankle, elbow, shoulder, and wrist. More than 90% of the cases are monarticular.

Hematogenous seeding, local spread, or traumatic or surgical infection may cause septic arthritis. In children, it most commonly results from hematogenous spread as bacteria pass into the synovial space through the highly vascular synovial membrane. The synovial membrane lacks a limiting basement membrane, which facilitates bacterial translocation. The bacteria then bind to bone and cartilage, initiating an inflammatory response that breaks down the joint by two mechanisms: directly through the effects of proteolytic enzymes and indirectly through pressure necrosis caused by accumulation of purulent synovial fluid.

Contiguous spread of infection from osteomyelitis to the joint space occurs in approximately 10% of cases and is more common in newborns and young infants. In these children, blood vessels cross the physis, thereby connecting the metaphysis and epiphysis and allowing bacteria direct access into the joint space. In addition, the joint capsules of the hip and shoulder overlie the bony metaphyses of the femur and humerus, facilitating direct extension of osteomyelitis into these joint spaces.

The most common bacterial causes of septic arthritis are listed in Table 176-5. Overall, Staphylococcus aureus is the most common cause of septic arthritis and infection in children; community-acquired methicillin-resistant S. aureus (MRSA) accounts for up to 65% of S. aureus infections.⁵² In neonates, group B streptococci and gram-negative organisms are common causes. In addition to the organisms listed in Table 176-5, additional causative organisms include Neisseria gonorrhoeae in neonates and sexually active adolescents, Pseudomonas aeruginosa and Candida species in injection drug abusers, Salmonella species in children with sickle cell disease, and gram-negative bacteria in immunosuppressed children. Kingella kingae, a fastidious gram-negative coccobacillus that colonizes the respiratory and oropharyngeal tracts in children, has been implicated as a common cause of osteoarticular infections in young children. K. kingae was found to be the causative agent in up to 50% of osteoarticular infections.^53,54 In previously well children, more than 95% of K. kingae infections are diagnosed between 6 months and 4 years of age; it may also cause infections in older children with underlying medical conditions.⁵⁵ Fortunately, K. kingae is susceptible to a wide array of antibiotics that are given empirically to young children for septic arthritis.

The clinical picture of septic arthritis varies with age. Infants tend to have fever, failure to feed, lethargy, pseudoparalysis of the extremity, and pain with being handled. In one study, however, neonates (younger than 1 month) were found to have less fever and fewer systemic signs of illness than in older infants, thus making the diagnosis even more difficult. Most older children have systemic signs and symptoms of fever, malaise, poor appetite, and irritability as well as localized symptoms of pain and refusal to move the affected joint or, if the lower part of the body is involved, limp or refusal to walk. With septic arthritis, the onset of symptoms is more acute than with osteomyelitis. Physical examination reveals local erythema, warmth, and swelling. If the hip is affected, it often is held in flexion, abduction, and external rotation. Range of motion is decreased because of pain and muscle spasm, and passive joint movement is painful. In infants, joint dislocation may be observed.

Laboratory studies helpful in diagnosis of septic arthritis include a white blood cell count, ESR, CRP assay, blood cultures, and evaluation of joint fluid. Use of the white blood cell count and ESR is discussed in the section on transient synovitis. The ESR rises 24 hours or more after the onset of signs and symptoms of infection, so it may not be helpful during the first day of illness. The CRP level may be a better monitor than the ESR for septic arthritis. The CRP assay is a simpler test that requires only a finger-stick sample of blood. The CRP level rises more quickly than the ESR, typically is elevated at the time of the initial evaluation, and with appropriate therapy will normalize within a week; by contrast, the ESR will not normalize for more than a month.⁵⁶ In patients with septic arthritis, the peripheral white blood cell count, ESR, and CRP level generally are elevated, although white blood cell counts and acute-phase reactants may be normal, especially with K. kingae infection.^55,57

Blood cultures should be included in the evaluation of a child with suspected septic arthritis. Culture results are positive in 20 to 50% of cases.⁵⁶ When culture results are positive, they not only help direct antibiotic treatment but also provide an organism for serum bactericidal testing as the child progresses to oral antibiotics.

If N. gonorrhoeae infection is suspected, special media are required for cultures of joint fluid, blood, pharynx, skin lesions, cervix, urethra, vagina, and rectum. Urine, urethral, cervical, and vaginal specimens also can be obtained for nucleic amplification testing.

If the patient has signs or symptoms of pharyngitis, a throat swab should be sent for culture of Streptococcus pyogenes. Antibody titers to antistreptolysin O and anti-DNase B also may help in establishment of a causative organism.

Evaluation of synovial fluid is the standard modality for diagnosis of septic arthritis. If septic arthritis is being considered, joint aspiration should be performed without delay and the sample sent for Gram’s stain, aerobic and anaerobic cultures, cell count with differential count, and glucose determination. The synovial fluid in patients with septic arthritis tends to be turbid or grossly purulent, with a white blood cell count higher than 40,000 cells/µL and a predominance of polymorphonuclear cells (Table 176-6). Synovial glucose concentration may be low (synovial fluid glucose/blood glucose ratio of less than 0.5) and protein and lactate levels elevated. Because of the intrinsic immunoglobulins in the synovial fluid, results on culture of the fluid will be positive in only half of the children in whom the clinical picture is consistent with septic arthritis. Joint fluid should be inoculated directly into blood culture bottles to enhance identification of fastidious organisms such as K. kingae.^52,57 Culture specimens may need to be incubated for a week or longer.⁵⁴

In septic arthritis of the hip, plain radiographs may be normal in appearance or, in the presence of a large joint effusion, may show periarticular soft tissue swelling, widening of the joint space, obliteration or displacement of the gluteal lines, and asymmetrical fullness of the iliopsoas and obturator soft tissue planes. Late in the course of infection, subchondral bone erosions and narrowing of the joint space are seen. Ultrasonography is much more sensitive than plain radiography for detection of hip effusion and provides direct visualization of the fluid and needle during joint aspiration. Scintigraphy also may be useful in diagnosis of septic arthritis; during the “blood pool” or delayed images of the joint, symmetrical uptake in periarticular tissue on both sides of the joint is seen. Scintigraphy is diagnostic of septic arthritis earlier than other imaging techniques are and is a useful adjunct in identification of associated osteomyelitis or avascular necrosis of the femoral head. CT can confirm the presence of an effusion but does not differentiate septic from nonseptic arthritis. Changes in signal intensity and femoral epiphysis perfusion seen on MRI may provide diagnostic clues in attempting to differentiate septic arthritis from transient synovitis.58,59

Septic arthritis requires immediate hospital admission, antibiotics, and surgical intervention. Surgical options range from needle aspiration to open surgical drainage, but no randomized, controlled trials have compared these two treatment approaches. Some experts recommend surgical drainage in all infants and young children with septic arthritis because needle aspiration has been shown to be inferior in this population.⁶⁰ Indications for surgical drainage in children with septic arthritis include involvement of the hip joint, presence of large amounts of pus or debris in the joint, loculated fluid, recurrence of joint fluid after four or five aspirations, and lack of clinical improvement within 3 days of the initiation of appropriate therapy.^61–64 In joints other than the hip, the need for surgical drainage is determined on a case-by-case basis.

Empirical antibiotic therapy for septic arthritis is directed against the most likely organisms as dictated by the patient’s age and comorbid conditions (see Table 176-5). Treatment may then be changed after culture and sensitivity results are known. If a pathogen is not isolated but the patient is improving, initial antibiotic therapy is continued. If a pathogen is not identified and the patient is not improving, reaspiration of the joint or the possibility of a noninfectious process should be considered. To maximize culture results, antibiotics should not be given until a specimen of joint fluid is obtained. Initial treatment should be parenteral to ensure adequate serum antibiotic concentrations. In neonates, the entire course of antibiotics should be administered parenterally. Patients older than 1 month can be transitioned to oral antibiotics after they have been afebrile for 48 to 72 hours and have improved signs and symptoms of infection, a white blood cell count that has normalized, and decreasing inflammatory markers (CRP and ESR). In general, doses two to three times those used for mild infections are sufficient. Response to therapy is measured by clinical improvement and assays for acute-phase reactants (CRP and ESR).

The mortality rate associated with septic arthritis has fallen to less than 1%, but the morbidity remains significant. Sequelae include leg-length discrepancy, persistent pain, limited range of motion and ambulation, and ischemic necrosis of the femoral head. Predictors of a poor outcome include infection of the hip or shoulder, adjacent osteomyelitis, delay of 4 days or more before antibiotic and surgical intervention, and prolonged time to sterilization of synovial fluid.

Legg-Calvé-Perthes Disease: Idiopathic avascular necrosis of the proximal femoral epiphysis, also known as Legg-Calvé-Perthes disease, is named after the men who independently described it in the early 1900s. It usually occurs between the ages of 3 and 12 years, with the peak incidence between the ages of 5 and 7 years. Legg-Calvé-Perthes disease has been reported in teenagers as well as in children as young as 2 years. Boys are affected three to five times more frequently than girls, and the disease is familial approximately 10% of the time. The disorder is bilateral in up to 20% of cases. Trauma often is related to the onset of symptoms, but a direct relationship between the two has not been clearly established. Thrombophilia also may be associated with Legg-Calvé-Perthes disease; however, studies are conflicting.

Despite extensive research, the etiology of Legg-Calvé-Perthes disease remains unclear. Ponseti and colleagues⁶⁵ suggested that the disease is a local manifestation of a transient generalized disorder of epiphyseal cartilage. Their histologic studies showed thickening and disorganization of the growth plate, which could block penetration of the blood vessels and render the femoral epiphysis avascular. Other etiologic theories involve abnormalities in the vascular anastomotic network around the femoral epiphysis, increased blood viscosity leading to infarction, and abnormalities in growth hormone.

On clinical examination, children with Legg-Calvé-Perthes disease initially exhibit a limp of insidious and stuttering onset. Associated pain, when present, usually is related to activity and relieved by rest. The pain tends to be localized to the groin or referred to the anteromedial aspect of the thigh or knee region. Symptoms and limp severity usually are worse at the end of the day. On examination, children have limited hip motion, particularly abduction and internal rotation. Early in the course of disease, the limited abduction is secondary to synovitis and muscle spasm. As the disease progresses, limitation of motion is due to deformity of the femoral head, and with time, the limited abduction may become permanent. Children with Legg-Calvé-Perthes disease demonstrate Trendelenburg’s sign (see the earlier section on developmental dysplasia of the hip) accompanied by thigh, calf, and buttock atrophy related to disuse. With advanced disease and femoral head collapse, limb-length discrepancy may be noted.

The role of laboratory evaluation in the diagnosis of Legg-Calvé-Perthes disease is limited to ruling out of other causes of hip pain (e.g., septic arthritis) when the diagnosis is in question or evaluation for hormonal, metabolic, or genetic causes in patients with bilateral hip involvement.

Legg-Calvé-Perthes disease is diagnosed and staged by plain radiographs obtained in the AP and frog-leg lateral views. This radiographic survey shows the extent of epiphyseal involvement and the stage of the disease and also provides prognostic information. Radiographic “head at risk” signs that are associated with poor results include a radiolucent V-shaped defect in the lateral epiphysis and adjacent metaphysis (Gage’s sign), speckled calcification lateral to the capital epiphysis, diffuse metaphyseal reaction (metaphyseal cysts), lateral capital femoral epiphysis subluxation, and horizontal physis. These prognostic signs help guide treatment. In early Legg-Calvé-Perthes disease, radionuclide bone scanning may be diagnostic before the development of abnormalities on plain film. Bone scintigraphy also has been shown to provide accurate information concerning the extent of the necrotic process as well as the degree of vascularization and hence the stage of the disease. Scintigraphy, however, was not able to predict disease outcome.⁶⁶ Compared with plain radiography and bone scanning, MRI gives earlier and more reliable information about the extent of necrosis of the femoral head. MRI also is better than scintigraphy at showing revascularization. Despite the usefulness of MRI in early Legg-Calvé-Perthes disease, its role in healed Legg-Calvé-Perthes disease is limited because it provides no additional information about the configuration and structure of the femoral head beyond that from plain radiography.⁶⁶ Arthrography is useful in delineation of flattening of the femoral head, demonstration of the hinge abduction phenomenon with abduction of the leg, and, in conjunction with plain films or CT, diagnosis of osteochondritis dissecans after Legg-Calvé-Perthes disease.

There are four radiographic classification stages of Legg-Calvé-Perthes disease: initial, fragmentation, reossification, and healed. Radiographic findings in the initial stage include a femoral head that appears smaller than the opposite unaffected femoral head, widening of the medial joint space, subchondral lucent zone (subchondral collapse, the so-called crescent sign; Fig. 176-25), irregular physeal plate, and blurry and radiolucent metaphysis. In the fragmentation phase, the repair aspects of the disease become more prominent. The epiphysis begins to fragment, and areas of increased radiolucency appear as new bone forms as well as areas of increased radiodensity. During the reossification stage, the repair process continues as normal bone density returns, radiodensities replace radiolucencies, and alterations in the shape of the femoral head and neck become apparent. The healed stage is the final radiographic stage of Legg-Calvé-Perthes disease, and radiographs of the proximal third of the femur obtained during this stage of the illness will demonstrate any residual deformities.

Multiple prognostic classification systems of Legg-Calvé-Perthes disease have been devised. In general, a poor prognosis is associated with a greater degree of deformity of the femoral head and acetabulum at maturity, disease onset in children 6 to 8 years of age or older, female gender, and prolonged duration of disease.

When the diagnosis of Legg-Calvé-Perthes disease is suspected, orthopedic consultation is recommended. Goals in the treatment of Legg-Calvé-Perthes disease are to improve range of motion, to prevent deformity, to limit growth disturbance, and to prevent degenerative joint disease. Treatment is not indicated in all affected children, but when it is indicated, it should be started in the initial or fragmentation phase of the disease. Treatment is recommended for patients who have a poor prognosis on the basis of clinical and radiographic findings. The cornerstone for treatment of Legg-Calvé-Perthes disease is referred to as containment: the femoral head is contained within the acetabulum to equalize pressure on the head and to subject it to the molding action of the acetabulum to maintain the spherical nature of the femoral head. Containment can be achieved through nonoperative or operative methods, and optimal treatment is determined on a case-by-case basis.

Residual deformities resulting from Legg-Calvé-Perthes disease follow four patterns: coxa magna (enlargement and deformation of the head of the femur), premature physeal arrest, irregular head formation, and osteochondritis dissecans. Despite these deformities, most patients with childhood hip disease are capable of bone remodeling with subsequent improvement in femoral head deformities. During their early years, regardless of radiographic appearance, patients with Legg-Calvé-Perthes disease tend to do well.⁶⁷ A majority (70-90%) of patients are active and pain free and have good range of motion 20 to 40 years after the onset of symptoms. Only patients with flattened irregular femoral heads at the time of primary healing or premature physeal closure experienced clinical deterioration and increasing pain.⁶⁸ Later in life, function becomes markedly reduced, and degenerative joint disease develops in a majority of patients by their 50s and 60s.⁶⁷ With noncontainment treatment, however, the outcome is not as good. Yrjonen found that at an average of 35 years of follow-up, 48% of patients had evidence of degenerative joint disease and 17% either had undergone total hip arthroplasty or had clinical symptoms that warranted total hip arthroplasty.⁶⁹

Slipped Capital Femoral Epiphysis: SCFE refers to posterior and inferior slippage of the proximal femoral epiphysis on the metaphysis. The femoral head sits securely in the acetabulum, whereas the epiphysis separates from the femoral neck through the growth plate. The average annual incidence of SCFE ranges from 0.2 (in Japan)⁷⁰ to 10.8 (in the United States) per 100,000 children; boys are affected almost twice as frequently as girls are.⁷¹ The peak incidence is during the adolescent growth spurt: boys between 12 and 16 years of age (mean age, 13.5 years) and girls between 10 and 14 years of age (mean age, 11.5 years). The age at diagnosis decreases with increasing obesity. A majority of children with SCFE have delayed skeletal maturation, with bone age being as much as 20 months behind chronologic age. African American and Hispanic children are affected more frequently than white children are. The literature reports SCFE to be bilateral in up to 80% of cases, although 30 to 40% of these cases are asymptomatic and discovered only on screening radiographs. In unilateral cases, the left hip is affected twice as often as the right.

SCFE is associated with endocrine disorders (hypothyroidism, panhypopituitarism, hypogonadism, and growth hormone administration), renal osteodystrophy, and radiation therapy. Most cases of SCFE, however, are idiopathic and associated with obesity. The etiology of idiopathic SCFE is unknown. It is likely to be multifactorial and related to biomechanical factors such as obesity and physeal architecture as well as hormonal factors that weaken physeal strength. Obesity results in increased shear forces across a more vertically and posteriorly oriented growth plate that has been weakened by architectural irregularities and hormonal changes of puberty. The consequence is slippage of the epiphysis inferiorly and posteriorly in the direction of the weight-bearing force.

The traditional classification of SCFE was based on the duration of symptoms, with acute slippage defined as symptom duration of less than 3 weeks, chronic slippage defined as symptoms for 3 weeks or longer, and acute-on-chronic slippage defined as symptom duration of more than 3 weeks with a recent sudden exacerbation. This classification has been replaced by one based on stability: in stable SCFE, ambulation is possible (with or without crutches), whereas in unstable SCFE, ambulation is not possible (with or without crutches).⁷² This classification system is preferred to the traditional classification because it does not rely on patient or parent recall for duration of symptoms and provides information about prognosis.

The signs and symptoms of SCFE vary with its stability. Children with stable SCFE have symptoms of intermittent limp and pain of several weeks to months in duration. Stable SCFE is found in approximately 95% of all cases.⁷² The pain of SCFE may be localized to the hip but more commonly is poorly localized to the thigh, groin, or knee. Atypical manifestations of SCFE include weakness and easy fatigability of the affected limb and limping on exertion. With continued slippage, internal rotation, flexion, and abduction are lost, and parents and children may note progressive external rotation and shortening of the involved lower extremity with subsequent difficulty in daily activities, such as tying shoes. On examination, children initially have a slight loss of internal rotation and experience pain only at the extremes of motion. Their gait is antalgic, and muscle atrophy is minimal. As the slip becomes more severe, the gait becomes more antalgic, internal rotation is lost, abduction and flexion of the hip increase, thigh and gluteal muscle atrophy is more pronounced, and leg-length discrepancy develops. A frequently seen sign associated with SCFE can be elicited during passive flexion of the affected hip: as flexion is increased from an extended position, the thigh abducts and externally rotates.

Unstable SCFE in children typically initially comes to attention after a sports-related injury or a fall with a twisting injury. These children experience acute onset of extreme pain and, on examination, hold the hip in flexion, external rotation, and abduction. The children resist any type of movement of the affected leg, and if unstable SCFE is suspected, no passive movement should be attempted for fear of further displacement of the epiphysis.

The diagnosis of SCFE is made with AP and lateral radiographs of both hips. With stable slippage, AP and frog-leg lateral pelvic radiographs should be obtained. When an unstable slip or a minimal slip is suspected, a cross-table radiograph replaces the frog-leg lateral view. Early in the course of SCFE, the initial slippage is posterior, so the AP view is normal in appearance or shows widening of the physis; the slip is better seen on the lateral projection (Fig. 176-26). Early findings on lateral radiographs include a minimal posterior step-off at the anterior physeal plate and widening of the growth plate. On AP radiographs, signs of slippage include Klein’s line and the blanch sign of Steele. Klein’s line is a line drawn along the superior margin of the femoral neck. With a normal hip, the line intersects with or falls within the epiphysis, whereas in a hip with a slipped epiphysis, the line does not come in contact with the epiphysis. The blanch sign of Steele is a crescent-shaped area of increased density in the proximal portion of the femoral neck that is created by superimposition of the posteriorly displaced epiphysis on the femoral neck. The femoral metaphysis of the affected hip also appears to be more laterally displaced from the medial wall of the acetabulum. As the slip continues, the epiphysis continues to be displaced inferiorly and posteriorly relative to the metaphysis. Over time, remodeling smoothes away the superior and anterior portions of the femoral neck, and in an attempt to buttress the slippage, callus is formed at the inferior and posterior portions of the femoral neck. Acute, unstable slips in children without previous symptoms will not be associated with evidence of bone healing or remodeling.

Ultrasonography has been used in the diagnosis of SCFE but offers little if any information beyond that provided by conventional radiography. CT and MRI can be used to confirm and to quantitate epiphyseal displacement in patients who have symptoms but no radiographic findings suggestive of SCFE.

Slip severity is described in either of two ways. The simplest classification describes the amount of displacement of the femoral head on the femoral neck. In mild SCFE, the amount of displacement is less than one third; moderate SCFE has displacement between one third and one half; and severe SCFE is marked by displacement of the femoral head by more than half the width of the femoral neck. A more accurate description of the magnitude of slippage is obtained by measurement of the epiphyseal-shaft angle of Southwick. On a frog-leg lateral radiograph of the pelvis, a line is drawn between the anterior and posterior tips of the epiphysis at the physeal plate level. A second line is then drawn perpendicular to the epiphyseal line. Next, a line is drawn along the midshaft of the femur. The epiphyseal-shaft angle is formed by the intersection of the perpendicular and the femoral shaft lines. The magnitude of slip displacement is the angle of the involved hip minus the angle of the normal hip. If involvement is bilateral, 12 degrees is used as the control angle. Mild SCFE involves displacement of less than 30 degrees, a moderate slip is between 30 and 50 degrees, and severe displacement is more than 50 degrees.⁷³

When SCFE is diagnosed in the ED, non-weightbearing status is assigned immediately, and the child is provided with either crutches or a wheelchair; an urgent orthopedic consultation should be obtained. For a stable slip, definitive treatment is best performed within a few days; however, hospital admission is prudent. With an unstable slip, some orthopedic surgeons recommend immediate fixation.

The goals of treatment are to prevent further slippage and to achieve physeal stability, to avoid complications, and to maintain adequate hip function. A stable slip can be fixed with internal fixation (pinning), bone graft epiphyseodesis, corrective osteotomy, or spica cast immobilization. Most orthopedic surgeons recommend internal fixation with a single central screw. Failure of this repair is rare, and with new techniques that avoid opening of the hip joint, it has become a procedure associated with minimal blood loss and wound complications and minimal need for hospitalization and rehabilitation. Postoperatively, non-weightbearing or touch-toe weightbearing for 4 to 6 weeks is followed by a gradual return to normal activity. Running and contact sports may be resumed once the physis is closed. With unstable SCFE, controversy has arisen about the timing of the surgery, the optimal number of fixative devices that should be used, whether reduction should be performed, and whether preliminary traction and bed rest should be used before surgery. Treatment is dependent on the degree of slippage and the preference and experience of the orthopedic surgeon.

The two most worrisome short-term complications of SCFE are avascular necrosis and chondrolysis; the risk for either process increases with the severity of the initial slippage.⁷⁴ Avascular necrosis has been reported to occur in 10 to 15% of children with SCFE. The risk is higher with unstable SCFE,^75,76 greater degrees of slippage, and multiple attempts at reduction.⁷⁶ In children with stable slips treated by single-screw internal fixation without reduction, reported rates of avascular necrosis are as low as 0 to 5%.⁷⁷

Chondrolysis occurs in approximately 5% of patients with SCFE, with an increased incidence in children with greater degrees of slippage. Chondrolysis may develop before and as a result of treatment. It should be suspected when pain and loss of motion are disproportionate to the severity of the slip. On radiographs, loss of articular cartilage may be seen. When SCFE is treated by spica cast immobilization, the rate of chondrolysis is reported to range from 19 to 67%.78,79 Approximately 50% of cases of chondrolysis will resolve during the course of several years; however, in other cases, it may progress to such severe pain and contractures that hip arthrodesis is needed.

Other complications of SCFE include nonunion, premature closure of the epiphyseal plate, and degenerative changes. Degenerative hip arthritis develops gradually during decades and has an earlier onset with more severe degrees of slippage.

Osgood-Schlatter Syndrome: In 1903, Osgood and Schlatter independently reported traumatically induced apophyseal injury to the tibial tubercle in adolescents.⁸² This entity, now known as Osgood-Schlatter syndrome, is the most common of the apophyseal disorders.⁸³ It is found most commonly in boys between 10 and 15 years of age and in girls between 8 and 13 years of age and frequently is bilateral. Boys are affected more often than girls are, although this trend is changing as more girls are becoming increasingly involved in competitive sports. Bilateral involvement is found in 20 to 30% of patients.

Clinically, patients with Osgood-Schlatter syndrome have tenderness, pain, and swelling at the site of insertion of the patellar tendon on the tibial tubercle. The tibial tubercle may be prominent and the quadriceps tight. Pain is worsened with activities (such as running and jumping) that cause the quadriceps to contract, thereby stressing the tubercle. Extension of the knee against resistance causes pain, but resisted straight-leg raises are painless. Osgood-Schlatter syndrome is a clinical diagnosis, and radiographs generally are not indicated. If radiographs are obtained, however, lateral views of the tibial tubercle may appear normal or may show an enlarged, fragmented, and irregular tibial tuberosity with or without an overlying bony ossicle. These findings are nonspecific, however, and are not diagnostic on their own. Ultrasonography, which has been proposed as a diagnostic tool by some experts, may reveal pretibial swelling, fragmentation of the ossification center, insertional thickening of the patellar tendon, or excessive fluid collection in the infrapatellar bursa.⁸⁴

Treatment of Osgood-Schlatter syndrome is conservative and symptomatic. Initially, the pain can be effectively managed with ice and modification of activity, with or without nonsteroidal anti-inflammatory medications. The use of a patellar strap may help relieve symptoms. After the acute inflammatory process has resolved, treatment focuses on strengthening and stretching of the quadriceps muscles. Mild pain during activity is not an absolute contraindication to participation; however, with more severe symptoms, the risk of avulsion of the tibial tubercle should be weighed against the benefits of competing. In rare instances, conservative therapy is insufficient, and a 2- to 3-week trial of crutches is required. Steroid injections have been used in the past but are not currently recommended owing to the associated risk of patellar tendon rupture. Complete recovery without residual pain or weakness is the rule. Recovery usually occurs within weeks but in some cases may not be complete until the underlying growth plate is closed. However, if the patient reaches skeletal maturity and is still symptomatic, consideration should be given to surgical removal of the tibial tubercle or the bony ossicle overlying the tibial tubercle (or both).

Sever’s Disease: Sever’s disease is an apophysitis of the calcaneus due to traction by the gastrocnemius-soleus complex. It initially was described in 1912⁸⁵ and commonly is manifested as posterior heel pain in an 8- to 13-year-old athlete. It is bilateral in 60% of cases. As with other apophyseal injuries, pain is exacerbated by activity. Sever’s disease can be associated with growth or tight heel cords or other biomechanical abnormalities. Impact sports, especially those that involve running, and sports in which cleats are worn frequently are implicated in the development of Sever’s disease.

Patients have pain at the insertion site of the Achilles tendon and plantar fascia on the calcaneus. Tenderness is elicited when the calcaneus is squeezed bilaterally. Dorsiflexion of the ankle is restricted by tight heel cords. Radiographs may be normal in appearance or may show partial fragmentation and increased density of the calcaneal apophysis, although these findings also can be seen in normal feet.

As in Osgood-Schlatter syndrome, treatment of Sever’s disease is conservative and symptomatic. Treatment consists of ice, massage, stretching of the plantar fascia and involved muscles (gastrocnemius-soleus complex and ankle invertors and evertors), nonsteroidal anti-inflammatory drugs, and addition of shock-absorbing shoe inserts. Heel cups also have been found to be helpful, but they should be accompanied by stretching to avoid exacerbation of the calf muscle contracture. Patients in whom conservative management fails have been found to have bone bruising and edema in the calcaneal metaphysis and apophysis, suggesting that calcaneal stress fractures may be the cause of persistent heel pain. Modification of activity and a trial of crutches for 3 to 4 weeks may be necessary.⁸⁶ Plantar fasciitis can cause heel pain as well, but patients will experience tenderness along the plantar fascia, especially at the attachment of the fascia to the calcaneus. Treatment involves the use of anti-inflammatory agents, rest, and relaxation of the plantar fascia with heel support pads.

Little League Elbow: Little League elbow is a term used to describe a group of injuries of the elbow, including apophysitis, medial epicondylitis, and osteochondritis dissecans of the radial head and capitellum. The injury involves an overuse phenomenon resulting in inflammation at the site where the forearm flexor muscles originate. As its name suggests, it commonly affects baseball pitchers in children’s leagues and is associated with overhead arm motion. It also is seen in other skeletally immature overhand athletes, including tennis players. Little League elbow is caused by the excessive forces placed on the medial side of the elbow during the late cocking and acceleration phases of throwing.

Patients tend to be preadolescents in whom pain on the medial aspect of the elbow is associated with diminished throwing effectiveness and decreased throwing distance. Examination reveals localized tenderness and swelling over the medial epicondyle and pain with resisted flexion of the wrist. There may be a slight flexion contracture. Radiographs may be normal in appearance or may show fragmentation, sclerosis, and widening of the medial epicondylar apophysis. Treatment consists of a regimen of ice, nonsteroidal anti-inflammatory drugs, and modification of activity. Throwing is restricted until the symptoms have resolved. After resolution of the pain, the patient can begin a program of muscle stretching and strengthening, with a gradual return to throwing. Recovery usually takes 4 to 6 weeks. If pain returns during resumption of pitching, rest should be reinstituted. Patients with persistent pain, locking or decreased range of motion of the joint, or lateral elbow pain should be evaluated for avulsion fractures, loose bodies, and osteochondritis dissecans.

To minimize the risk of medial epicondylitis in children’s league pitchers, the American Academy of Pediatrics recommends limiting the number of pitches to 200 per week or 90 pitches per outing⁸⁷; however, the USA Baseball Medical and Safety Advisory Committee recommends lower pitch counts: 75 to 125 pitches per week or 50 to 75 pitches per outing, depending on age.⁸⁸ Preseason conditioning, use of proper pitching mechanics, and gradual increases in the amount and intensity of throwing are other preventive measures.

Apophysitis of the Hip: Apophyseal injuries of the hip involve sites around the hip where major abdominal and hip muscles either originate or insert: the anterior superior iliac spine, the anterior inferior iliac spine, the iliac crests, and the ischial tuberosities. Dancers and distance runners most commonly are affected. Patients with apophysitis of the hip experience dull pain near the hip that is related to activity. Treatment includes strengthening and stretching of the abdominal and hip muscles and restriction of activity.

Avulsion Fractures: Apophysitis tends to have an insidious onset; therefore, any patient with sudden onset of apophyseal pain after an acute traumatic event should be evaluated for an avulsion fracture of the apophysis or the adjacent bone. Such fractures occur when the muscle attachments to the apophyses are pulled off during strong active contractions against resistance, such as with a single powerful throw. Examination will reveal localized tenderness and swelling. The diagnosis usually is readily apparent on plain films (Fig. 176-27); however, bone scan and ultrasonography may play a role when the diagnosis is in question. Ultrasonography is advantageous because it involves no radiation exposure, it can detect a fracture before the development of an ossification center, and it provides a dynamic examination. Findings consistent with an avulsion fracture include a hypoechogenic zone, increased distance to the apophysis, dislocation of the apophysis, and mobility of the apophysis on dynamic examination.⁸⁹ Treatment of an avulsion fracture is based on the degree of separation. If the displacement is minimal (less than 2 cm with a hip avulsion and less than 5 mm with avulsion of the medial epicondyle⁹⁰), immobilization for 4 to 6 weeks with subsequent slow resumption of activities usually is sufficient. With widely separated avulsion fractures, some experts recommend open reduction and fixation.⁹⁰