The concepts of skeletal system injuries in children are quite different from those of adults. Children at different stages of development have varying responses of the skeletal system to trauma. Most of the anatomic and physiological features of the pediatric skeletal system result in a diminished propensity for serious skeletal injuries in comparison to adults.

The muscles, tendons, and ligaments of children are more resilient and forgiving of injury than are those of adults. The forces associated with accidents in young children are usually not great, in part because the leverage is reduced by the relatively small size of the extremities. Another important factor is the variation in bone composition at different ages. The presence of growth centers in the child also modifies the skeletal response to injury. Injuries to the immature skeleton carry the important potential for subsequent growth disturbances.

The skeleton of the young child is more pliable than that of adults, principally because the Haversian canals at this age are relatively large and contain more water. A proportionately lower mineral content in growing bone than in the mature skeleton is an additional, but less important, factor. These properties impart greater elasticity and plasticity to the bone.

The periosteum is thicker, more elastic, and less firmly attached to the bone in children than in adults. When injured, the periosteum of the child is likely to remain intact and attached to the fracture fragments. The periosteum provides a hinge-like mechanism between the fracture fragments and thereby aids in fracture reduction and stability; fracture nonunion is rare in children.

The changes in activity that tend to occur at different ages with normal growth and development greatly influence the types of skeletal injuries that occur in children. Child abuse is an important cause of skeletal injuries in infants and toddlers. Falls are the most common mechanisms of injury in ambulatory children. Outstretched arms usually dampen falls; this diminishes the impact to the head, but increases the likelihood of injuries to the elbow and forearm. The radius is the most commonly fractured bone in children. Clavicle fractures are also common, usually due to the transmission of force from an impact against the shoulder. Although falls most often lead to injuries of the upper extremity, exceptions occur when the child falls backward with substantial force. These falls carry a greater risk for skull fractures and intracranial injuries. In infants, intentional injury is an important cause of skull fractures.

Bicycle accidents, skateboard injuries, contact sports injuries, motor vehicle crashes, and pedestrian accidents are common in adolescents and older children. The types of injuries that are associated with these mechanisms of trauma are much like those identified in adults. Many sports activities are associated with relatively specific types of musculoskeletal injuries in the immature skeleton.

Young children are uniquely susceptible to incomplete fractures, that is, those that do not involve the entire width of the bone. These are termed greenstick fractures, torus fractures, and bowing injuries (plastic deformation).

The greenstick fracture is the result of a bending or angulation stress that places the convex side of the bone in tension and the concave side in compression. This process is similar to the bending of a green twig. A greenstick fracture refers to cortical failure in response to the tension forces along the convex side. The fracture is usually in the mid-diaphysis and involves about half the circumference of the shaft. The peripheral portion of the fracture is usually perpendicular to the shaft of the bone, and the central component tends to occur as a longitudinal split that can extend in 1 or both directions. The cortex along the concave surface remains intact, but is bowed. The most common sites of greenstick fractures are the radius, ulna, clavicle, and fibula. Radiographs of a greenstick fracture show angulation or bowing of the bone and an incomplete fracture line in the cortex of the convex side (Figure 65-1).

A torus fracture is a cortical buckle injury that is due to a compression force. This fracture typically occurs in the metaphysis of a long bone. The degree of buckling is variable. Some of these fractures are difficult to detect on radiographs. Careful inspection of technically adequate images is essential. In general, less force is required to produce a torus fracture than a greenstick fracture. The most common locations of torus fractures are the distal portions of the radius and ulna, the proximal aspect of the tibia, and the base of the first metatarsal. The distal radius is the most frequent site (Figure 65-2). A true torus fracture is a buckle fracture in which the cortex buckles externally. A fracture in which the cortex buckles internally (concave) is best termed a buckle fracture.

A torus or buckle fracture involves disruption of the normal smooth architecture of the cortex. The pathophysiology relates to microfractures that allow cortical buckling, often without a radiolucent line on radiographs. There is interruption of the normal smooth cortical margin. Fragment separation does not occur. Bone impaction at the fracture site usually results in subtle increase in density on radiographs. In some instances, a transverse radiolucent fracture line extends from the site of cortical buckling (Figure 65-3). Torus fractures usually heal within 2 to 3 weeks. Long-term sequelae are rare, as angulation is usually minimal, and there is no growth plate involvement.

A lead pipe fracture is a combination of an incomplete transverse fracture of 1 side of the cortex (i.e., a greenstick fracture) and a torus fracture of the opposite side (Figure 65-4). The osseous changes that occur with this injury are analogous to the response of a lead pipe to bending. Lead pipe fractures most often occur in the same skeletal sites as torus fractures.

Angulation or compressive forces applied to the long bones of infants and children sometimes result in a bending or bowing deformity without a radiographically visible fracture line. Typically, the deformity involves the entire length of the bone. This injury is termed a bowing injury (fracture) or plastic deformation. These fractures are often radiographically subtle. A comparison view of the contralateral structure is sometimes helpful for determining if minor bowing is pathologic. Common sites of plastic deformation fractures are the radius, ulna, and (occasionally) the fibula. The radius and ulna often bow concomitantly, or plastic deformation of 1 bone is associated with a complete or incomplete fracture of the companion bone. Plastic deformation of the fibula is usually associated with a complete tibial fracture.

Plastic deformation injury is a manifestation of the stress–strain curve phenomenon. The forces exceed the range of elastic strain of the bone that allows recovery of the normal shape, but are less than those necessary to produce a macroscopic fracture. Histologic examination shows a series of oblique microfractures on the concave side of the bowing deformity. A thin margin of periosteal new bone formation usually develops within several days of the injury, most prominently along the concave side (Figure 65-5). If there is severe deformation of the radius or ulna, restricted pronation and supination of the forearm can occur. Severe plastic deformation of the fibula sometimes interferes with proper alignment and apposition of an associated tibial fracture.

A common lower extremity injury in young children (approximately 9 months to 6 years of age) is the toddler fracture. This is a nondisplaced spiral fracture of the mid and distal portions of the tibial diaphysis, most often occurring due to the application of rotational force to the tibia as the child falls forward with the foot fixed. This injury is particularly common among infants who are just learning to walk, due to a broad-based in-toed gait caused by normal developmental tibial torsion at this age. In older toddlers, impaction fractures of the proximal tibial metaphysis can occur when the child jumps from a height. Toddler-type injuries can also occur in the small bones of the foot and ankle, including the calcaneus, navicular, and cuboid (Figure 65-6).

Toddler fractures of the tibia are usually nondisplaced and can be subtle on radiographs (Figure 65-7). Most are visible as a thin steep oblique radiolucency in the distal diaphysis on the anteroposterior (AP) view. Occasionally, the fracture orientation is such that the lateral view is diagnostic. On the lateral view, care must be exercised to avoid mistaking the nutrient canal for a fracture. An internal oblique view is often helpful for detecting a spiral tibial fracture when standard frontal and lateral radiographs are normal or equivocal. Bone scintigraphy is diagnostic in those cases in which there is a radiographically occult injury (Figure 65-8). Follow-up radiographs show periosteal new bone formation along the tibial diaphysis (Figure 65-9).¹

The epiphyseal complex includes the physis and the adjacent portions of the epiphysis and metaphysis. At least 15% of fractures in children involve these areas of growing bone. The epiphyseal complex is responsible for longitudinal bone growth, and has an important impact on the contour of the bone and the axial joint relationships. A fracture though the physis in a child is the pediatric analog of a dislocation or ligamentous injury in an adult. The ligaments and tendons are stronger than the cartilaginous growth plate.

An apophysis is an epiphyseal complex-equivalent structure that does not contribute to linear bone growth, but rather alters the shape of the bone at the site of attachment of tendons and ligaments. Because the muscles, tendons, and ligaments are stronger than the cartilaginous growth plate of the apophysis, avulsive stress tends to cause separation through the growth plate and displacement of the ossified portion of the apophysis in the direction of the contiguous tendon.

The growth plate (physis) of the epiphyseal complex or apophysis consists of 4 zones: (1) a zone of small germinal cells adjacent to the ossification center; (2) a zone of flattened cells arranged in columns, termed the proliferating cells; (3) a zone of hypertrophic cells, which are enlarged vacuolated cells that maintain the columnar arrangement of zone 2; and (4) the zone of provisional calcification, which is located adjacent to the metaphysis (some authors consider this to be a subdivision of the hypertrophic zone). The cartilage cells are in a matrix of chondroitin sulfate, through which longitudinally arranged lines of collagen are deposited. The collagenous fibers in the matrix resist shearing forces. There are abundant collagenous fibers in the first 2 zones, but they are less pronounced in the third (hypertrophic) zone because of enlargement of the cartilaginous cells. In zone 4, the matrix undergoes calcification.

The first 2 layers of the physis are relatively strong, due to the abundant collagenous matrix. Calcification in the matrix of the zone of provisional calcification adds strength to this layer. The nonmineralized portion of the hypertrophic zone (the third layer) is the weakest portion of the growth plate because of the limited amount of matrix and the lack of calcification. Most shearing injuries of the physis, therefore, involve the hypertrophic zone. Portions of the fracture line frequently involve the junction of the hypertrophic zone and the zone of provisional calcification. The germinal cells usually remain viable, and subsequent growth is normal, as long as there is no substantial vascular compromise.

The blood supplies of the metaphysis and epiphysis are independent prior to growth plate closure. Periosteal vessels and small branches from the nutrient artery supply the metaphysis. The periosteal vessels originate from the periphery of the diaphysis. The vascular supply of the epiphyseal ossification center consists of periosteal vessels that arise from the periarticular arteries. This dual vascularity is such that 1 set of vessels supplies nourishment to the epiphysis and physis for growth, whereas the metaphyseal arteries supply nourishment for ossification. Therefore, a fracture through the growth plate does not significantly interfere with the blood supply to either the epiphysis or the metaphysis.

An important component of the reparative response to a physeal fracture is deposition of fibrin within the line of separation. The cartilage cells continue to proliferate, and the epiphyseal plate widens. Resorption of the fibrin occurs over the next few weeks, and normal growth resumes. No clinically detectable growth disturbance occurs with most physeal fractures.

Angular deformities or bone shortening are important, but infrequent, potential sequelae of physeal fractures. Growth plate fractures of the proximal femur and the radial head have a greater propensity for deformity in comparison to most other fractures. The anatomy at these sites is unique in that the epiphyses are completely intracapsular, that is, they are surrounded by the joint capsule and are covered by articular cartilage. At least a portion of the epiphyseal arterial supply at these sites must enter from the metaphyseal side. These vessels are in the perichondrium at the peripheral margin of the growth plate and are susceptible to disruption with a physeal fracture. Avascular necrosis of the growth plate and epiphysis can follow such injuries, potentially leading to deformity of the epiphysis and arrest of longitudinal growth.

The Salter-Harris system is a method for classification of epiphyseal complex fractures (Figure 65-10). A Salter-Harris type 1 fracture is a pure epiphyseal separation. A shearing or avulsive force produces a fracture that is limited to the zone of hypertrophied cartilage of the physis. The germinal cells of the physis remain intact. This fracture does not directly involve the metaphysis or the epiphyseal ossification center. Type 1 fractures account for approximately 5% of epiphyseal complex injuries. These injuries are sometimes radiographically occult, and the diagnosis is based on clinical findings alone; the distal portion of the fibula is the most common site of a radiographically occult Salter-Harris type 1 fracture. The radiographic detection of a Salter-Harris type 1 fracture is based on the demonstration of malalignment between the epiphysis and metaphysis and/or widening of the growth plate (Figure 65-11). Follow-up radiographs usually show minimal adjacent periosteal new bone formation as well as bone resorption along the margins of the physis.

Young children are most susceptible to Salter-Harris type 1 injuries because of the relatively wide morphology of the physis. Most children with this fracture are younger than 5 years of age. Salter-Harris type 1 fractures can occur as a result of a traumatic birth. The wide and deficiently ossified physes of children with rickets or scurvy are prone to pathological type 1 fractures. Epiphysiolysis of the proximal femoral epiphyseal ossification center (slipped capital femoral epiphysis) or humeral head is a clinically severe form of a Salter-Harris type 1 injury (Figure 65-12). Epiphysiolysis can occur in child abuse (Figure 65-13).

The Salter-Harris type 2 fracture courses through a portion of the growth plate and then passes through the peripheral aspect of the metaphysis (Figure 65-14). A roughly triangular fragment from the metaphysis accompanies the displaced epiphyseal ossification center. As in the type 1 injury, the line of separation in the physis is limited to the hypertrophied zone. The Salter-Harris type 2 fracture accounts for more than 75% of epiphyseal complex injuries in childhood. The distal radius is the site of approximately 40% of these fractures; the distal tibia, distal fibula, phalanges, metacarpals, distal femur, and distal ulna follow in decreasing order of frequency. Salter-Harris type 2 fractures are common throughout childhood, until growth plate closure. The peak ages are between 12 and 15 years.

Intraarticular shearing forces cause Salter-Harris type 3 fractures. The fracture begins on the articular surface and courses vertically through the epiphyseal ossification center (entering the hypertrophic layer of the growth plate) and then horizontally (to the peripheral aspect of the physis). The metaphysis is not involved. Radiographs show a vertically-oriented epiphyseal fracture line and widening of the involved portion of the physis (Figure 65-15). This results in detachment of a portion of the epiphysis, but the severity of displacement is variable and in many patients is minimal. In some instances, there is no visible physeal widening. A nondisplaced type 3 fracture is sometimes radiographically subtle and is only visible on views oriented perpendicular to the fracture line (Figure 65-16). The most common sites of type 3 fractures are the distal tibia, the phalanges, and, occasionally, the proximal tibia (Figure 65-17). This injury accounts for less than 10% of epiphyseal complex injuries. Growth arrest and deformities are rare. The fragment must be adequately reduced, however, to restore the contiguity of the articular surface. Improper or incomplete reduction carries the potential of joint surface irregularity that can lead to degenerative joint disease.

A Salter-Harris type 4 fracture is due to a vertically oriented splitting force that produces a linear fracture through the epiphysis, growth plate, and metaphysis (Figure 65-18). A fragment composed of portions of the metaphysis and the epiphysis is usually separated from the parent bone. Type 4 injuries account for less than 10% of growth plate fractures. The most common sites are the lateral condyle of the humerus in children younger than 10 years of age and the distal portion of the tibia in older children. Growth arrest and joint deformities can occur with this fracture, although they are minimized by proper open reduction and fixation. Because the articular cartilage is involved in the Salter-Harris type 4 fracture, accurate reduction is essential to restore the contour of the articular surface.

The Salter-Harris type 5 injury is rare, accounting for only approximately 1% of growth plate injuries. This injury consists of physeal damage from a crushing mechanism. Initially, radiographs are normal or show subtle narrowing of the physis. The effects of the injury may not be apparent until sometime later, manifesting as bone and joint deformities. The distal femoral and distal tibial growth plates are the most common sites of Salter-Harris type 5 fractures. This injury tends to involve slightly older children than the other forms of epiphyseal complex injuries; the peak age range is 12 to 16 years. Motor vehicle incidents and athletic trauma account for most of these injuries.

Various additions and modifications to the classic Salter-Harris classification system are sometimes utilized. The type 6 injury involves the peripheral region of the physis (affecting the perichondral structures) that may occur due to an avulsion or glancing-type of trauma. The small fracture fragment of a type 6 injury usually consists of growth plate, metaphysis, and epiphysis. Type 7 is an isolated injury of the epiphysis, without physeal involvement. An osteochondral fracture is a type 7 injury. A type 8 injury refers to an isolated injury of the subphyseal metaphysis, without growth plate involvement. Type 9 is an injury of the periosteum and subperiosteal cortex.²

Approximately 80% of epiphyseal complex injuries are due to shearing or avulsive stress. This can occur as a pure shearing force applied directly to the epiphyseal center or as an avulsion force transmitted through attached ligaments or the joint capsule. The orientation of the force vector is parallel to or away from the epiphyseal ossification center. Apophyseal separations are nearly always the result of shearing or avulsive injuries.

The remaining 20% of epiphyseal complex injuries are due to splitting or compression forces that are oriented vertically with respect to the epiphysis and are transmitted through the apposing bones at the joint. This may create a split in the bone that crosses the entire epiphyseal complex perpendicular to the growth plate. This mechanism of injury is most often associated with fractures of the distal portions of the humerus and tibia. With healing, callous often extends across the growth plate at the fracture site, potentially fixing the epiphyseal ossification center to the metaphysis and creating a localized growth arrest. The remainder of the epiphyseal complex continues to grow, leading to a variable degree of angular deformity.

A compression force can also manifest as a crushing injury of the physis (such as a type 5 injury) that is accompanied by a variable degree of resting cell destruction. This fracture, although radiographically subtle, is associated with a substantial risk of growth arrest and deformity. The distal femur and distal tibia are the most common sites of this injury.

At least 80% of epiphyseal complex injuries occur between the ages of 10 and 16 years. A major exception to this pattern is fractures of the distal humerus, which usually occur before the age of 10 years. The age of peak occurrence of epiphyseal complex injuries is somewhat younger for girls (8 to 13 years) than for boys (11 to 14 years). Overall, these injuries are more common in boys. This is probably due to the greater exposure of boys to trauma and the relative delay of growth plate closure in boys because of slower skeletal maturation.

The most common site of epiphyseal complex injuries in children is the distal radius; this location accounts for 60% of all growth plate injuries. Approximately 20% of epiphyseal complex injuries occur in the distal tibia and 10% in the phalanges of the hand. In general, fractures of the epiphyseal complex are more common in the distal epiphyses of the long bones than in the proximal epiphyses.

MR serves an important role for selected patients with a known or suspected growth plate injury (Figure 65-19). This is particularly useful in young children for evaluation of unossified cartilage.

MR provides accurate assessment of a suspected transcondylar fracture in the young child with a predominantly cartilaginous epiphysis. A fracture typically appears as a linear hypointense defect in cartilage and bone on T1-weighted images. Edema adjacent to the fracture is hyperintense on short tau inversion recovery (STIR) and T2-weighted fat-suppressed sequences. On T2-weighted sequences, a transverse physeal fracture appears as a linear defect that separates the high signal intensity physeal cartilage from the low signal intensity zone of provisional calcification. Fat-suppressed gadolinium-enhanced T1-weighted images are sometimes helpful, as physeal cartilage enhances markedly, whereas the fracture space is avascular. In patients with a chronic physeal injury, as occurs with gymnasts, MRI shows physeal irregularity and widening. There may be vertical extensions of physeal cartilage into the metaphysis and thin transphyseal bridges.³

A nondisplaced Salter-Harris type 3 fracture is sometimes difficult to detect with standard radiographs; there may also be confusion with nonpathologic developmental epiphyseal clefts. A type 3 fracture is sometimes visible on only 1 projection. CT and MR are highly accurate for the detection of these vertical epiphyseal fractures. Scintigraphy shows abnormal increased tracer accumulation. Scintigraphy is also helpful for the diagnosis of a Salter-Harris type 1 fracture in selected patients.

Apophyseal Fractures

An apophysis is a bony prominence to which muscles or tendons attach. In the immature skeleton, a physis is located at the junction between an apophysis and the parent bone. Most apophyseal injuries result from avulsive forces transmitted through muscles and tendons. This can occur as a single traumatic event or repetitive microtrauma, that is, traction apophysitis. Common sites of traction apophysitis in children include the knee (Osgood-Schlatter disease), heel (Sever disease), medial epicondyle of the humerus (Little League elbow), and ischial tuberosity.

Apophyseal separation fractures are common in the pelvis and hips. The single most frequent location is the ischial apophysis and the adjacent portion of the ischium, which is the site of attachment of the hamstring muscles. Other potential pelvic sites of avulsion injuries include the anterior superior iliac spine, the anterior inferior iliac spine, the iliac crest, and the lesser trochanter of the femur. In the knee, acute fractures of the tibial tubercle can occur due to avulsive force applied by the quadriceps muscles via the patellar ligament. The medial humeral epicondyle is the most common upper extremity site of an apophyseal fracture. The apophyses of the olecranon process of the ulna and the coracoid and acromion processes of the scapula are additional sites at which these injuries occasionally occur.

There are several characteristic avulsion injuries of the pelvis and hips that occur in young athletes. These injuries are particularly common among track athletes. The muscles involved in these avulsion injuries are as follows: (1) sartorius muscle and tensor fascia lata—anterior superior iliac spine; (2) rectus femoris muscle—anterior inferior iliac spine and along a groove just above the superior aspect of the acetabular rim; (3) psoas major muscle—lesser trochanter (strenuous hip flexion); (4) hamstrings—the apophysis of the ischial tuberosity (hurdlers); (5) gluteal muscles—the greater trochanter; (6) abdominal musculature—the apophysis of the iliac crest (associated with abrupt directional change during running); (7) adductor muscles—injuries in the region of the symphysis pubis; and (8) the reflected head of the rectus femoris—the acetabular rim.

The typical clinical manifestations of an avulsive injury include focal pain, tenderness, and swelling. Radiographs show displacement of the ossified portion of the apophysis (Figure 65-20). There usually is irregularity at the surfaces of the separation (Figure 65-21). Skeletal scintigraphy shows focal increased tracer accumulation. CT imaging is sometimes helpful for more accurate depiction of the pathological anatomy or for documentation of a subtle fracture (Figure 65-22). On T2-weighted or proton density–weighted fat-suppressed MR images, an acute avulsive injury leads to hyperintense edema within the marrow and soft tissues in the region of the fracture (Figure 65-23). There may also be edema and hemorrhage in the involved muscle. Potential findings on T1-weighted images include laxity of the attached tendon, a hypointense fracture line, and displacement of an osseous fracture fragment. Follow-up radiographs may show new bone formation or healing with incorporation of the avulsed fragments into the parent bone. Occasionally, irregular new bone formation at the site of an avulsive injury results in an imaging appearance that simulates a bone neoplasm.

Traction Apophysitis

Overuse injuries involving an apophysis are common in children. Traction apophysitis represents repetitive microtrauma of an apophysis due to the transmission of force through an attached muscle and tendon. This mechanism is distinct to that of an acute apophyseal avulsion fracture, which is due to a single traumatic event. Patients with traction apophysitis usually experience long-standing, sometimes progressive, pain at the site of involvement. Physical examination demonstrates focal tenderness and soft tissue swelling. In the lower extremities, the findings are often bilateral. The most common site of traction apophysitis is the tibial tubercle, that is, Osgood-Schlatter disease. Other potential sites include the medial epicondyle, the olecranon, the posterior aspect of the calcaneus (Sever disease), at the base of the fifth metatarsal (Iseling disease), and the ischial apophysis.

The radiographic appearance of traction apophysitis consists of physeal widening and apophyseal irregularity and sclerosis. The radiographic diagnosis of traction apophysitis is complicated by normal variation in apophyseal ossification. Correlation with the clinical findings is essential when there is an equivocal radiographic appearance. MR demonstrates marrow edema in the apophysis, edema and widening of the physis, and edema in the adjacent soft tissues. There is focally increased uptake in the apophysis on skeletal scintigraphy.

Traumatic Enthesopathy

An enthesis is the osseous insertion site of a tendon, ligament, or joint capsule. Enthesopathy is a nonspecific term that refers to pathology at an enthesis. Inflammatory enthesopathy can occur in children with a systemic arthritis such as juvenile idiopathic arthritis. Repetitive avulsive forces at an enthesis can lead to stress-related traumatic enthesopathy. Radiography and CT may show reactive bone formation or demineralization at the site. With active inflammation, MR demonstrates edema in the soft tissues and bone.⁴

Stress Fractures

Musculoskeletal overuse injuries occur due to repetitive application of submaximal stresses, rather than a single acute event. Overuse injuries are becoming more common due to increasing participation of children in rigorous organized sports. These injuries can involve bone, articular cartilage, or growth cartilage (i.e., the epiphyseal and apophyseal growth plates), with the latter tissue having the greatest degree of susceptibility. Anatomic factors that can predispose a child to overuse injuries include leg length discrepancy, hip or long bone malalignment, and foot deformities.

A stress fracture is an overuse injury of the bone, in which a focal osseous disruption results from the repeated application of stress. This is in contradistinction to most other fractures, in which a single acute event causes abrupt loss of the structural integrity of the bone. Stress fractures can result from compression or distraction forces. They most often occur at sites subjected to repetitive muscular pull. A stress fracture probably begins as a microscopic focus of cortical bone infraction that progresses with continued activity. Periosteal new bone formation develops as a reactive phenomenon.

There are 2 types of stress fracture. (1) The fatigue fracture is due to the repeated application of abnormal stresses to normal bone. Fatigue fractures may develop over a period of days to weeks or, less commonly, after severe physical exertion of only several hours duration. (2) An insufficiency fracture results from normal stresses applied to abnormal bone. Conditions associated with insufficiency fractures in children include juvenile idiopathic arthritis, osteogenesis imperfecta, rickets, fibrous dysplasia, and osteoporosis of any cause.

A stress fracture typically causes local pain that is relieved by rest and accentuated by physical activity. The onset of pain may be abrupt or gradual. There is usually a history of a repetitive strenuous activity, such as running or gymnastics. Stress fractures can occur at sites of muscular attachment. The tibia is the most frequently involved bone in children, most often along the posterior portion of the upper half of the diaphysis. Additional common sites of stress fractures include the metatarsals (“march fracture”), fibula, calcaneus, tarsal navicular, patella, femur, ribs, pelvis, spine, humerus, ulna, and foot sesamoids.

Because most stress fractures take some time to develop, radiographically visible periosteal new bone formation is often present at the time of initial evaluation. In some cases, there is a transverse line of sclerosis in the cortex, due to compression of trabeculae or reactive cancellous new bone formation (Figure 65-24). A transverse linear radiolucent fracture line in the cortex can also occur (Figure 65-25). Subtle cortical abnormalities that are sometimes visible on radiographs or CT prior to an overt fracture include focal osteopenia (“gray cortex”), a small round or oval intracortical resorption cavity, and intracortical linear lucencies (cortical striation). Because of limited sensitivity for marrow edema, normal radiography or CT does not exclude the diagnosis of a clinically significant stress injury.^5–7

MRI is the most sensitive imaging technique for the detection of stress injuries. The most important MR finding is edema in the marrow at the site of the injury, with high signal on STIR images and relatively low signal intensity on T1-weighted sequences (Figure 65-26). When a fracture line is present, it is slightly hyperintense to cortical bone and hypointense to marrow fat (Figure 65-27). Reactive sclerosis along the fracture line is hypointense on all imaging sequences. Cortical thickening, when present, is also hypointense. Periostitis results in edema adjacent to the cortex. On STIR or fat-suppressed T2-weighted images, the elevated periosteum is sometimes visible as a thin hypointense membrane surrounded by hyperintense edema.⁸

Scintigraphy is also sensitive for the detection of stress fractures, even in the early stages (Figure 65-28). Most often, skeletal scintigraphy in these patients shows a sharply marginated area of increased tracer uptake. In other patients, the imaging pattern consists of ill-defined increased uptake involving a longer segment of a long bone; there may also be a combination of diffuse and focal patterns (Figure 65-29). In young athletes, scintigraphy commonly shows additional, asymptomatic, sites of stress change within the extremity contralateral to the symptomatic lesion or at other areas in the ipsilateral extremity.⁹

Medial tibial stress syndrome (shin splints) is a common cause of lower leg pain in runners and other young athletes. Most clinicians define this condition as exercise-induced lower leg pain without imaging evidence of a focal stress fracture. The most common cause is inflammation at the origins of the soleus and flexor digitorum longus muscles. Predisposing factors include hindfoot varus, excessive forefoot pronation, genu valgum, and excessive femoral anteversion. Medial tibial stress syndrome is rare in children below the age of 15 years. Physical examination demonstrates tenderness along the surface of the mid to distal portion of the tibia. The tenderness is usually more diffuse than is typical for a stress fracture.

Radiographs of patients with medial tibial stress syndrome are often normal. Periosteal elevation or subtle cortical indistinctness occurs in some patients. With chronic stress, cortical thickening can occur. Bone scintigraphy often shows increased uptake along the posteromedial aspect of the tibia. The most common MR findings are subperiosteal and marrow hyperintensity on STIR or fat-suppressed T2-weighted images (Figure 65-30). Some patients with clinical manifestations of medial tibial stress syndrome have normal imaging studies. Also, stress-related MR signal alterations in the tibial marrow can occur in athletes who have no lower leg symptoms.¹⁰

Spondylolysis

Spondylolysis is a defect in the pars interarticularis, most often occurring as a stress fracture from repetitive trauma. This is a relatively common cause of persistent low back pain in adolescents. Ballet dancers, gymnasts, and football players are at an elevated risk for spondylolysis. Spondylolysis can usually be detected on high-quality oblique radiographs as a linear radiolucency in the pars interarticularis. As with other stress injuries, the radiographic findings may be subtle or absent early in the course of the process. Bone scintigraphy, CT, or MRI can be utilized for earlier detection or to confirm findings demonstrated on standard radiographs (Figure 65-31). CT is the most sensitive imaging technique for detection of a pars defect. An acute lesion is often thin and lacks substantial reactive sclerosis (Figure 65-32). An older lesion has reactive sclerosis adjacent to the defect. Occasionally, cortical bone covers the margins of the defect. A pars defect appears as a hypointense band on MR. Edematous marrow is hyperintense on STIR and fat-suppressed T2-weighted sequences; sclerotic bone is hypointense (Figure 65-33).

In patients with unilateral spondylolysis, stress in the contralateral pars can cause pain. This is termed Wilkinson syndrome. Scintigraphy shows prominent uptake in the stressed pars. CT shows sclerosis; thickening, irregularity, and small lucencies are sometimes present as well (Figure 65-34). MR demonstrates edema and/or sclerosis.

Conservative treatment options for spondylolysis include restriction of activities, immobilization, and physical therapy. A prompt diagnosis with scintigraphy or MR sometimes allows early institution of treatment and prevents progression to an overt pars defect. Information from bone scintigraphy or MR also helps guide therapy in selected patients with spondylolysis by documenting the intensity of the reparative process. In some patients, there is a persistent radiographic pars defect despite fibrous stabilization.

Osteochondritis Dissecans

Osteochondritis dissecans involves separation of a segment of subchondral bone, usually with secondary abnormalities of the overlying articular cartilage. This condition is divided into juvenile and adult types, based on whether the adjacent growth plate is fused. The juvenile form has a better prognosis, and surgical therapy is less frequently indicated, than in the adult form. The mean age at presentation for juvenile osteochondritis dissecans is approximately 12 years. The lateral aspect of the medial femoral condyle is the most common location. Other potential sites include the elbow (capitellum, radial head), ankle (superomedial aspect of the talar dome), humeral head, and patella. Approximately 60% of affected individuals are male.^11,12

The pathophysiologic mechanism of juvenile osteochondritis dissecans is multifactorial. Repetitive trauma appears to be an important initiating factor, and many patients have a history of participation in sports. The initial lesion is likely a subchondral stress fracture. One or more bone fragments become separated from the adjacent normal epiphyseal bone and undergo ischemic necrosis. Additional fragmentation of the ischemic bone may occur. Intraarticular loose bodies in these patients can occur due to delamination of the overlying cartilage or displacement of an ischemic bone fragment.¹³

Most patients with osteochondritis dissecans present with nonspecific pain, sometimes aggravated by vigorous activity. Physical examination may demonstrate focal tenderness and soft tissue swelling. Findings that suggest the presence of a loose interarticular fragment include crepitance, popping, and locking. Treatment of a stable lesion includes immobilization and non-weight bearing; persistent or separated lesions can be treated arthroscopically. Resolution of juvenile osteochondritis dissecans occurs with activity restriction alone in 80% to 90% of patients.

The integrity of the overlying articular cartilage is an important factor in the clinical course of osteochondritis dissecans. Early in the process, the cartilage is usually intact, but may have secondary changes such as swelling or thinning. If the cartilage remains intact, healing of the bone lesion usually occurs. Healing in the early stages can occur with revascularization of the fragment and new bone formation at the sites of separation. In the later stages, dead fragments of bone are resorbed as revascularization occurs, and viable fragments grow to fill the defect. If the weakened cartilage overlying the lesion tears, spontaneous healing is unlikely. One or more osteochondral fragments may be extruded into the joint when the cartilage is torn. Osteochondritis dissecans can be classified arthroscopically as follows: grade I, irregularity and softening of overlying cartilage, without a fissure; grade II, articular cartilage tear, without fragment displacement; grade III, displaceable, partially attached fragment; and grade IV, displaced osteochondral fragment into the joint.¹²

The radiographic appearance of osteochondritis dissecans is that of a radiolucent defect in subchondral bone (Figure 65-35). There is usually an irregular, round or oval osseous fragment within the defect. When a fragment is present, a radiolucent zone separates the fragment from the adjacent bone. A sclerotic zone is present in the adjacent viable bone. The fragment sometimes appears radiodense relative to normal bone, because of ischemia. In other patients, there is no visible fragment associated with the subchondral radiolucent defect; this either occurs due to necrosis and resorption of the fragment or displacement of the fragment into the joint space (Figure 65-36). Intermediate forms are common, with partial fragment resorption and multiple small irregular fragments.

The defect of osteochondritis dissecans is sometimes difficult to visualize on standard radiographs. Oblique views are sometimes useful. In the knee, a flexed or tunnel view may best demonstrate the lesion (Figure 65-37). The most common location of osteochondritis dissecans of the knee is in the posterolateral aspect of the medial femoral condyle.

CT imaging is helpful in selected patients with osteochondritis dissecans to accurately delineate the location and extent of the lesion (Figure 65-38). Standard CT or CT arthrography allows detection of an intraarticular loose body and assessment of the status of the overlying articular cartilage.

MRI effectively demonstrates the pathological anatomy of osteochondritis dissecans. The MR findings are helpful in selecting those patients who would most likely benefit from surgical intervention.¹¹ The earliest manifestation of osteochondritis dissecans with MR is a focus of abnormal signal in the subchondral epiphyseal bone and swelling of the overlying articular cartilage. The osseous lesion is hypointense on T1-weighted images; the signal intensity is variable on T2-weighted sequences, but is often predominantly hyperintense on fat-suppressed T2-weighted images. Next, bone fragmentation is identifiable and there is a linear band of high signal intensity on T2-weighted images at the interface between the normal and abnormal bone (Figure 65-39). The bone fragment at this stage is hypointense on all imaging sequences. Mild signal alteration due to edema is present in the adjacent marrow in some patients. Bone cysts can sometimes be identified as round foci of high signal intensity on T2-weighted images. The next temporal change is thinning of the adjacent articular cartilage, sometimes with small focal areas of altered signal intensity. The lesion is “stable” up to this point, as the articular cartilage is intact, and healing can occur with conservative therapy. Healing is indicated by coalescence and growth of bone fragments to fill the subchondral defect and return of normal thickness and signal intensity of the articular cartilage.

Progression of osteochondritis dissecans to an unstable lesion (arthroscopic grades II to IV) that requires surgical intervention is uncommon in children. An unstable lesion is indicated on MR by a tear in the articular cartilage or a displaced osteochondral fragment (Figure 65-40). In the absence of a displaced fragment, the differentiation between a stable and an unstable lesion is often difficult. The most reliable sign with MR is a fluid-filled gap in the articular surface, which is bright on T2-weighted images and dark on T1-weighted images. Some investigators have stressed the importance of a high signal intensity interface on T2-weighted images between the lesion and the adjacent normal bone as indicating extension of synovial fluid around the fragment, and thereby confirming an unstable lesion. However, this MR appearance can also occur in stable lesions due to cystic changes or granulation tissue. MR arthrography can be helpful in equivocal cases.^14–16

The imaging appearance of osteochondritis dissecans overlaps that of various other entities, such as osteonecrosis, osteochondral fracture, multiple epiphyseal dysplasia, and accessory ossification centers. There is usually a history of an acute injury in patients with an osteochondral fracture. Involvement of multiple epiphyses is typical of epiphyseal dysplasia and osteonecrosis. However, individuals with multiple epiphyseal dysplasia or Stickler dysplasia may have a predilection for developing osteochondritis dissecans or avascular necrosis of the epiphyses.

Normal developmental irregularities of the femoral condyles are common in young children. These may occur as multiple ossification centers, subchondral radiolucent defects, or spiculated condylar margins. Developmental irregularities most commonly are located in the posterior portions of the condyles and often involve both knees. On radiographs, adjacent sclerosis is minimal or absent. MR shows intact overlying cartilage and lack of edema in the cartilage; the radiolucent zones correspond to uncalcified cartilage on MR. In the absence of superimposed stress changes, marrow edema is lacking. These features aid in the differentiation from stage I osteochondritis dissecans. The common occurrence of spiculation or accessory ossifications in the posterior-inferior central aspects of the condyles with normal-appearing articular cartilage is a normal developmental variation in ossification. The presence of a central intracondylar lesion with surrounding edema on MR indicates true osteochondritis dissecans. There is typically a thin hyperintense zone between the lesion and the normal articular cartilage on T2-weighted MR images. In some instances, however, there are equivocal diagnostic imaging findings. Sequential MR examinations are sometimes helpful in this situation, usually in conjunction with conservative therapy.

Shoulder Impingement Syndrome

In the upper extremities, both the shoulder and elbow are relatively common sites for overuse injuries. The shoulder internal impingement syndrome is often related to athletic activities that involve throwing. During extreme abduction and external rotation, the posterior fibers of the supraspinatus tendon and/or the anterior fibers of the infraspinatus tendon can be impinged between the humeral head and the posterior aspect of the glenoid. Radiographs of these patients are normal. Potential findings on MR include undersurface tears of the supraspinatus or infraspinatus tendons, traumatic cysts in the posterior aspect of the humeral head, and injuries of the posterosuperior labrum.¹⁷

Little League Elbow

Little League elbow is an overuse syndrome that is associated with repetitive throwing or other athletic activity involving the arm. This most often occurs as osteochondrosis of the capitellum (Panner disease). Other potential manifestations of this type of overuse include osteochondritis dissecans of the capitellum, trochlea, or radial head, epiphysiolysis or stress changes of the medial epicondyle and olecranon, and ulnar collateral ligament injury. Patients report pain localized to the area of involvement and loss of range of motion.

Early in the course, standard radiographs of Little League elbow are normal or show nonspecific soft tissue changes. Eventually, alterations in morphology and radiodensity of the involved osseous structures occur. Common findings in the medial epicondyle include enlargement and sclerosis, separation from the underlying bone, and fragmentation. Involvement of the capitellum, radial head, or trochlea is suggested by focal subchondral destruction and sclerosis. MR and sonography allow confirmation of the diagnosis early in the process while radiographs are still normal. The involved bones are hyperintense on fat-suppressed T2-weighted images (Figure 65-41). In young children prior to growth plate closure, MR may show widening and prominent T2 signal in the medial epicondyle growth plate and edema in the epicondyle (or hypointense sclerosis, if chronic). T2 hyperintensity in the medial collateral ligament indicates an associated tear.^18,19

Little League Shoulder

Little League shoulder, or proximal humeral epiphyseolysis, is an overuse injury of the proximal humeral physis that occurs in adolescent athletes who participate in sports that involve repetitive forceful overhead activity. Most affected patients have a history of intense competitive baseball pitching. Stress changes in the proximal humeral physis can also occur in gymnasts. There is gradual onset of proximal arm pain. Typically, there is tenderness to palpation over the proximal aspect of the humerus.

The most important radiographic finding of Little League shoulder is widening of the lateral aspect of the proximal humeral physis (Figure 65-42). There is usually slight irregularity at the physeal margins. The pathologic anatomy is best demonstrated on radiographs obtained in external rotation. On MR, the features include lateral physeal widening and marrow signal abnormality in the adjacent portion of the metaphysis (hypointensity on T1-weighted images and hyperintensity on T2-weighted images) (Figure 65-43). Occasionally, there are signal abnormalities in the epiphysis.^20–22

Gymnast Wrist

Gymnast wrist is an overuse injury of the growth plate of the distal aspect of the radius caused by repetitive compression forces. Patients report pain and swelling at the site. There may be limitation of dorsiflexion at the wrist. Initially, radiographs are normal or show minimal soft tissue changes. Eventually, widening and irregularity of the growth plate occur (Figure 65-44). Potential MR features include physeal widening, metaphyseal marrow edema, and extension of physeal cartilage into the metaphysis.²³

Overuse Injuries of the Knee

Patellofemoral Stress Syndrome

Patellofemoral stress syndrome (patellar compression syndrome) is probably the most common cause of knee pain in adolescents. It is an overuse injury of the articular cartilage. Lower extremity deformities such as patellar subluxation, patella alta, lateral patellar tilt, or excessive genu valgus can predispose to this condition or exacerbate the severity of symptoms. There are no specific radiographic findings of patellofemoral stress syndrome. MR may show signs of edema in the articular cartilage.

Growth Plate Stress

Competitive athletes can develop stress-related injuries of the growth plates of the knees. As with other overuse injuries, there is gradual onset of chronic pain. Radiographs demonstrate focal growth plate widening. This finding is also present on MR. Signal alteration in the adjacent portion of the metaphysis is common. In some patients, interference with normal physeal growth results in varus or valgus deformity.²⁴

Osgood-Schlatter Disease

Traction apophysitis at the insertion of the patellar tendon with the tibial tubercle (tuberosity) is termed Osgood-Schlatter disease (Osgood-Schlatter lesion). Repetitive traction at the tibial tubercle apophysis results in irritation and pain. Microtears occur in the deep fibers of the patellar tendon. Repetitive stress causes edema in the marrow of the tibial tubercle and in the apophyseal cartilage.^25,26

Children with Osgood-Schlatter disease report activity-related pain at the site of the tibial tuberosity. Physical examination demonstrates swelling and localized tenderness at this site. Pain is exacerbated by extension of the knee against resistance. The clinical presentation is usually around the age of puberty; the peak age is 13 to 14 years for boys and 10 to 11 years for girls.

The tibial tubercle develops as an inferior extension of the proximal tibial epiphysis. Ossification of the tibial tuberosity is first visible during late childhood or early adolescence. Irregular ossification and/or multiple ossification centers are commonly present during development of the tubercle.

Radiographs of Osgood-Schlatter disease show local soft tissue swelling anterior to the tibial tubercle, often accompanied by irregularity in ossification of the tubercle (Figure 65-45). The margins of the adjacent portion of the patellar ligament are ill-defined on the lateral radiograph, due to edema. If there is sufficient ossification of the tibial tubercle, the lateral radiograph often demonstrates avulsed bone fragments from the anterior inferior surface. There is considerable variation in the regularity of ossification of the tubercle in normal patients; therefore, attention to the adjacent soft tissues and correlation with the clinical findings are important.²⁷

The sonographic features of Osgood-Schlatter disease include cartilage swelling and edema, fragmentation of the ossification center, thickening of the patellar tendon, and bursitis of the infrapatellar bursa.²⁸

MRI of Osgood-Schlatter disease demonstrates hypertrophy and fragmentation of the tibial tubercle. Occasionally, there is ossification within the patellar tendon. These findings are usually best demonstrated on T1-weighted sequences. Fat-suppressed T2-weighted images show hyperintensity (due to edema) in the tibial tubercle, marrow of the adjacent epiphysis, inferior aspect of the patellar tendon, and adjacent subcutaneous fat (Figure 65-46). Fluid often accumulates in the deep infrapatellar bursa. Edema is occasionally present in the Hoffa fat pad. The inferior aspect of the patellar tendon is often thickened.^26,29

Sinding-Larsen-Johansson Syndrome

Sinding-Larsen-Johansson syndrome refers to a repetitive stress injury of the inferior aspect of the patella at the attachment of the patellar tendon. The peak age range for this overuse injury is 10 to 13 years. Radiographs may show irregular calcification along the inferior margin of the patella, but this does not occur until symptoms have been present for some time. With long-standing involvement, the inferior pole of the patella may elongate. The earliest finding is soft tissue swelling within and adjacent to the superior aspect of the patellar tendon. Sonography can be useful to demonstrate the soft tissue changes in these patients: irregularity of inferior patellar cartilage, edema of the patellar tendon, and fluid in the infrapatellar bursa. MR is the optimal technique to detect these soft tissue alterations. Fat-suppressed T2-weighted images usually demonstrate hyperintense edema in the inferior pole of the patella and the adjacent aspect of the patellar tendon (Figure 65-47).³⁰

Jumper’s Knee

Patellar tendinosis (jumper’s knee) is an overuse syndrome that is clinically similar to Sinding-Larsen-Johansson syndrome and Osgood-Schlatter disease except that there is no bone involvement. It typically occurs in young adult athletes; it is uncommon in skeletally immature children. Radiographs are normal or show soft tissue prominence in the region of the patellar tendon. MR demonstrates enlargement of the patellar tendon, usually most prominent proximally. Abnormally increased signal intensity is sometimes present within the tendon on STIR images.

Bipartite Patella

Bipartite patella is a developmental failure of fusion of patellar ossification centers. The prevalence in the general population is about 2%. It is much more common in males than in females. The anomaly is bilateral in about 40% of affected individuals. A great majority of individuals with bipartite patella have no associated symptoms, and the finding is a developmental variation. Rarely, the lesion is painful. Most patients with pain have a history of repetitive strenuous extension or, less often, direct trauma.

Radiography of bipartite patella shows a radiolucent curvilinear cartilaginous cleft that results in a separate patellar ossicle. The Saupe classification system recognizes three types: type 1 (5%) is at the inferior pole of the patella, type 2 (20%) is at the lateral margin, and type 3 (75%) is in the superolateral aspect. In patients with patellar pain, stress views occasionally demonstrate widening of the cleft. MR of an injured or stressed bipartite patella shows marrow edema at the margins. The chondroosseous junction may be irregular. Prominent uptake on bone scintigraphy is usually present in these symptomatic patients as well. The typical uncomplicated bipartite patella has normal uptake on scintigraphy and normal marrow signal intensity on MR.

Iliotibial Band Syndrome

Iliotibial band syndrome is an overuse disorder in which soft tissue inflammation results from repetitive friction between the inferior aspect of the iliotibial band and the lateral femoral condyle. This is the most common cause of lateral knee pain in long-distance runners. Cyclists are also prone to this condition. MRI shows soft tissue edema between the iliotibial band and the lateral femoral condyle (Figure 65-48). Fluid and inflammatory synovial thickening are usually present in the lateral synovial recess of the knee. With more severe involvement, there is edema in the adjacent portion of the lateral femoral condyle. Occasionally, the iliotibial band is thickened or edematous.^31,32

Osteochondritis Dissecans

The knee is the most common site of osteochondritis dissecans (see the discussion earlier in this chapter). This is a separation of a portion of subchondral bone from the adjacent epiphysis, often accompanied by alterations in the overlying articular cartilage. The posterolateral aspect of the medial femoral condyle is the most common location. The radiographic appearance of osteochondritis dissecans is that of a radiolucent defect in subchondral bone (Figure 65-37). There is usually an irregular, round, or oval osseous fragment within the defect.

The most common causes of an intraarticular loose body in children are osteochondritis dissecans and acute fractures. Synovial osteochondromatosis is an additional potential cause. Loose bodies can be cartilaginous, osseous, or osteocartilaginous. Most are laminated round osteochondral objects. The elbow is the most common location. Epiphyseal fractures and avulsion fractures of the medial epicondyle can lead to elbow loose bodies as well.

Loose bodies composed only of cartilage are not visible on radiographs. MR and CT or MR arthrography provide greater sensitivity. Long-standing osteochondral fragments usually are round and have a laminated appearance. The source of the loose body is often visible as a radiolucent defect in the articular cortex. Because the imaging evaluation often occurs some time after the osteochondral fragment separation in patients with osteochondritis dissecans, interval healing may result in a defect that is much smaller than the loose body. On arthrography, some loose bodies have a synovial attachment. Adjacent synovial thickening may be visible on MR. T2* gradient-recalled echo (GRE) images are useful for detecting small calcified loose bodies on MR. Larger lesions often have similar signal characteristics as normal bone on T1-weighted sequences. Loose bodies are hypointense relative to joint fluid on T2-weighted images. Fat-suppressed proton density images are best for demonstrating chondral fragments.

Synovial osteochondromatosis is the presence of multiple intraarticular or bursal osteocartilaginous bodies due to synovial metaplasia. Typically, only 1 joint is involved. The most common sites are the knee, elbow, hip, and shoulder. The most specific radiographic appearance is the presence of multiple calcified or ossified intraarticular bodies in the absence of signs of arthritis. The bodies are usually of similar size and character. The cartilaginous bodies contain hyalin cartilage that produces low to intermediate signal intensity on T1-weighted images and high signal intensity on T2-weighted sequences. Calcified bodies are hypointense on T1-and T2-weighted sequences. The MR signal characteristics of ossified lesions are similar to those of normal bone, with marrow signal intensity centrally and cortical signal peripherally. The intraarticular or bursal calcified bodies can also be detected with CT or sonography.^33–35

A pathologic fracture is one that occurs through bone that has been weakened by a focal lesion (e.g., bone tumor or infection) or a generalized or polyostotic skeletal process (e.g., bone dysplasia, Langerhans cell histiocytosis, metastatic disease, or rickets) (Figure 65-49). The most common lesion that is associated with pathologic fractures in children is a simple (unicameral) bone cyst (Figure 65-50). With a comminuted fracture of a simple bone cyst, a displaced fracture fragment sometimes moves to the dependent portion of the cyst, that is, the “fallen fragment sign.” Nonossifying fibroma is occasionally associated with complete or incomplete pathologic fractures in children. It can also impart susceptibility for an avulsion injury at the site of a tendinous insertion onto the thin cortex of the lesion. Occasionally, a stress fracture pattern occurs, with periosteal thickening along the cortex overlying the nonossifying fibroma.

Pathologic fractures can occur in any of various diffuse skeletal processes that lead to bone weakening. Vertebral body compression injuries are common in children with osteoporosis or marrow diseases, such as neurological impairment, leukemia, sickle cell disease, osteogenesis imperfecta, steroid therapy, and metastatic neuroblastoma. Multiple recurring fractures occur throughout the skeletal system in most children with osteogenesis imperfecta. Very low birth weight infants, particularly those with rickets, may suffer fractures of ribs (most common site) and other bones (radius > humerus > ulna > clavicle).³⁶ Conditions that predispose to traumatic epiphyseal separations include renal osteodystrophy, rickets, myelomeningocele, and septic arthritis.

A pathologic fracture and subsequent pseudarthrosis can occur in developmentally abnormal bone. The classic form of this phenomenon is in the tibia in patients with neurofibromatosis type 1. About half of children with pseudarthrosis of a long tubular bone have other manifestations of neurofibromatosis type 1, with the finding occurring as an isolated clinical entity in the other half (see Chapter 58). The pathophysiology of developmental pseudarthrosis typically involves deossification and/or thinning of a long bone, with subsequent bowing and fracture. There is subsequent lack of normal healing; callus formation is deficient. Fibroosseous tissue forms adjacent to the fracture site as a reactive phenomenon. The most common site of idiopathic and neurofibromatosis-related pseudarthrosis is the distal portion of the tibia; other long bones can also be involved, including the fibula, radius, and ulna. Pseudarthrosis of the forearm can occur in association with fibrous dysplasia.

The most common injuries sustained by neonates during vaginal delivery are clavicular fractures and brachial plexus injuries. There is an elevated risk for these complications in patients with a history of fetal macrosomia. Brachial plexus paresis occurs in 0.019% to 0.25% of livebirths. Most of these infants have substantial improvement with conservative therapy. The prevalence of clavicle fractures during childbirth is approximately 3 per 1000 livebirths; this injury accounts for at least 85% of all obstetric fractures. Nearly all birth-related clavicle fractures heal without sequelae.^37,38

Epiphyseal separations can occur during childbirth. The usual sites include the proximal and distal portions of the humerus, and the proximal femur; the distal femur, distal tibia, and fibula are additional potential sites. High birth weight infants of diabetic mothers who undergo a difficult vaginal delivery are at particular risk. Epiphyseal separation of the proximal humerus is usually related to hyperextension and excessive external rotation of the arm during vaginal delivery.

Fat necrosis results from aseptic saponification of fat by lipase released from blood and tissue. The most common precipitating events are blunt trauma and surgery. Fat necrosis can occur at any location; the breast is a common site. Fat necrosis can occur in the scrotum. Birth trauma can precipitate fat necrosis in young infants. Potential presentations of fat necrosis include a painless lump, tenderness, skin induration, ecchymosis, and skin thickening.

Sonography of fat necrosis demonstrates a lesion in the subcutaneous fat. In some patients, the lesion is isoechoic or hypoechoic and has a well-defined hypoechoic halo, typically with an oval shape (Figure 65-51). In others, there is a poorly defined hyperechoic focus in the subcutaneous fat. Acoustic enhancement is typically lacking. Color Doppler evaluation sometimes demonstrates prominent perfusion of the lesion. The most characteristic MR appearance is that of an oval subcutaneous lesion that contains internal fat lobules, septations, and a pseudocapsule. There is diffuse or ring-like enhancement. Inflammatory edema and enhancement in the adjacent subcutaneous fat are common. Other patients have a linear focus of signal alteration and enhancement in the subcutaneous tissues. CT demonstrates a fat-containing subcutaneous mass.^39,40

Muscle injuries are common in children, particularly in association with athletic activities. The pathologic spectrum in these patients includes muscle strain, contusion, tear, hematoma, and herniation. Mechanisms of injury include overuse, a direct blow, and rapid excessive muscle stretching. Muscle injuries also occur with other forms of trauma such as penetrating injuries. When clinically indicated, MR provides specific information concerning the nature and severity of muscle injuries.